The effect of dehorning and other potential factors influencing horn growth in game-ranched white rhinoceroses (Ceratotherium simum simum)
This is a notebook for the dehorning data of Ververs et al. (2018). When you execute code within the notebook, the results appear beneath the code.
Currently the following R packages were loaded
## Loading required package: knitr
## Loading required package: effects
## Loading required package: carData
## lattice theme set by effectsTheme()
## See ?effectsTheme for details.
## Loading required package: psych
## Loading required package: lsmeans
## The 'lsmeans' package is being deprecated.
## Users are encouraged to switch to 'emmeans'.
## See help('transition') for more information, including how
## to convert 'lsmeans' objects and scripts to work with 'emmeans'.
## Loading required package: FactoMineR
## Loading required package: factoextra
## Loading required package: ggplot2
##
## Attaching package: 'ggplot2'
## The following objects are masked from 'package:psych':
##
## %+%, alpha
## Welcome! Related Books: `Practical Guide To Cluster Analysis in R` at https://goo.gl/13EFCZ
The data was read from the excel file containing all dehornings, and some
This has a first larger dataset is loaded in memory to start from.
A summary is requested holding descriptive statistics.
summary(df.cv.final)## GENDER AH_W_G PH_W_G SHAVINGS_W_G
## F :1297 Min. : 5 Min. : 2.0 Min. : 6.0
## M : 747 1st Qu.: 849 1st Qu.: 142.0 1st Qu.: 61.0
## NA's: 1 Median :1263 Median : 259.0 Median : 99.0
## Mean :1339 Mean : 308.9 Mean :103.9
## 3rd Qu.:1670 3rd Qu.: 406.0 3rd Qu.:139.0
## Max. :5900 Max. :2336.0 Max. :495.0
## NA's :4 NA's :216 NA's :175
## TOTAL AH_L AH_C AH_STUMP_L
## Min. : 5 Min. : 15.0 Min. : 55.0 Min. : 10.00
## 1st Qu.:1017 1st Qu.:119.0 1st Qu.:505.0 1st Qu.: 75.00
## Median :1610 Median :145.0 Median :590.0 Median : 80.00
## Mean :1708 Mean :168.8 Mean :572.7 Mean : 84.74
## 3rd Qu.:2173 3rd Qu.:185.0 3rd Qu.:645.0 3rd Qu.: 90.00
## Max. :8166 Max. :835.0 Max. :870.0 Max. :145.00
## NA's :37 NA's :90 NA's :185
## PH_L PH_C PH_STUMP_L CALC_AGE
## Min. : 5.0 Min. : 20.0 Min. : 5.00 Min. : 58
## 1st Qu.: 40.0 1st Qu.:420.0 1st Qu.: 20.00 1st Qu.: 924
## Median : 50.0 Median :510.0 Median : 25.00 Median :1379
## Mean : 56.6 Mean :491.2 Mean : 27.25 Mean :1555
## 3rd Qu.: 65.0 3rd Qu.:575.0 3rd Qu.: 30.00 3rd Qu.:1967
## Max. :250.0 Max. :960.0 Max. :100.00 Max. :5909
## NA's :222 NA's :129 NA's :343 NA's :956
## CALC_ARRIVAL DH_YEAR DH_MONTH DH_QUARTER W_DIFF
## Min. :-318.0 2014 :548 11 :316 1:283 Min. :-498
## 1st Qu.: 479.0 2015 :531 9 :309 2:553 1st Qu.: 796
## Median : 786.0 2013 :340 5 :251 3:542 Median :1050
## Mean : 882.2 2012 :319 6 :195 4:667 Mean :1126
## 3rd Qu.:1274.5 2011 :252 12 :192 3rd Qu.:1335
## Max. :5468.0 2010 : 44 7 :167 Max. :5313
## NA's :498 (Other): 11 (Other):615 NA's :220
## L_DIFF C_DIFF B_YEAR B_MONTH B_QUARTER
## Min. :-106.0 Min. :-840.00 2010 :201 3 :169 1 :400
## 1st Qu.: 70.0 1st Qu.: 50.00 2008 :191 4 :124 2 :313
## Median : 90.0 Median : 80.00 2011 :190 1 :116 3 :183
## Mean : 113.4 Mean : 85.44 2012 :137 6 :116 4 :193
## 3rd Qu.: 135.0 3rd Qu.: 120.00 2009 : 98 2 :115 NA's:956
## Max. : 630.0 Max. : 608.00 (Other):272 (Other):449
## NA's :226 NA's :134 NA's :956 NA's :956
## DIM_DEHORN INT_DEHORN CALC_PARITY C_YEAR
## Min. : 1.0 Min. : 0.0 0 :1447 2011 : 132
## 1st Qu.: 285.8 1st Qu.: 453.0 1 : 356 2012 : 122
## Median : 487.0 Median : 587.0 2 : 187 2013 : 114
## Mean : 546.3 Mean : 589.9 3 : 47 2014 : 99
## 3rd Qu.: 721.5 3rd Qu.: 715.0 4 : 6 2010 : 79
## Max. :2070.0 Max. :1335.0 NA's: 2 (Other): 50
## NA's :1449 NA's :1060 NA's :1449
## C_MONTH C_QUARTER N_DEHORNINGS GR_DAYS
## 2 : 94 1 : 258 1 :1058 Min. : 0.0
## 3 : 93 2 : 152 2 : 648 1st Qu.: 540.0
## 4 : 82 3 : 120 3 : 284 Median : 692.0
## 1 : 71 4 : 66 4 : 48 Mean : 832.9
## 12 : 66 NA's:1449 5 : 5 3rd Qu.: 914.8
## (Other): 190 NA's: 2 Max. :4892.0
## NA's :1449 NA's :463
## G_TOTAL G_AH_W G_AH_L
## Min. : 0.0090 Min. : 0.0090 Min. : 0.0145
## 1st Qu.: 0.9284 1st Qu.: 0.8188 1st Qu.: 0.1728
## Median : 2.1526 Median : 1.7057 Median : 0.2165
## Mean : 3.0904 Mean : 2.3942 Mean : 0.2515
## 3rd Qu.: 3.2135 3rd Qu.: 2.4829 3rd Qu.: 0.2612
## Max. :332.6667 Max. :246.3333 Max. :18.3333
## NA's :465 NA's :469 NA's :491
## G_AH_C G_PH_W G_PH_L G_PH_C
## Min. : 0.0731 Min. : 0.0098 Min. : 0.0073 Min. : 0.0195
## 1st Qu.: 0.5434 1st Qu.: 0.1119 1st Qu.: 0.0456 1st Qu.: 0.4062
## Median : 0.8126 Median : 0.3198 Median : 0.0669 Median : 0.6739
## Mean : 1.0452 Mean : 0.6213 Mean : 0.0881 Mean : 0.8797
## 3rd Qu.: 1.0843 3rd Qu.: 0.5513 3rd Qu.: 0.0900 3rd Qu.: 0.9430
## Max. :111.6667 Max. :86.3333 Max. :10.0000 Max. :86.6667
## NA's :522 NA's :660 NA's :659 NA's :550
describe(df.cv.final)## vars n mean sd median trimmed mad min
## GENDER* 1 2044 1.37 0.48 1.00 1.33 0.00 1.00
## AH_W_G 2 2041 1338.94 738.70 1263.00 1267.16 609.35 5.00
## PH_W_G 3 1829 308.93 246.84 259.00 275.47 192.74 2.00
## SHAVINGS_W_G 4 1870 103.89 57.64 99.00 99.87 57.82 6.00
## TOTAL 5 2045 1707.62 984.53 1610.00 1616.93 856.94 5.00
## AH_L 6 2008 168.81 91.64 145.00 151.88 51.89 15.00
## AH_C 7 1955 572.67 108.74 590.00 578.41 96.37 55.00
## AH_STUMP_L 8 1860 84.74 14.57 80.00 83.94 14.83 10.00
## PH_L 9 1823 56.60 30.79 50.00 51.62 22.24 5.00
## PH_C 10 1916 491.24 127.77 510.00 498.75 103.78 20.00
## PH_STUMP_L 11 1702 27.25 10.09 25.00 26.70 7.41 5.00
## CALC_AGE 12 1089 1554.74 839.96 1379.00 1452.76 748.71 58.00
## CALC_ARRIVAL 13 1547 882.22 612.32 786.00 843.47 590.07 -318.00
## DH_YEAR* 14 2045 4.34 1.45 5.00 4.44 1.48 1.00
## DH_MONTH* 15 2045 7.48 3.22 8.00 7.64 4.45 1.00
## DH_QUARTER* 16 2045 2.78 1.05 3.00 2.85 1.48 1.00
## W_DIFF 17 1825 1125.98 514.80 1050.00 1067.21 398.82 -498.00
## L_DIFF 18 1819 113.44 73.81 90.00 101.23 44.48 -106.00
## C_DIFF 19 1911 85.44 62.89 80.00 84.73 59.30 -840.00
## B_YEAR* 20 1089 10.55 2.49 11.00 10.76 2.97 1.00
## B_MONTH* 21 1089 5.58 3.45 5.00 5.37 4.45 1.00
## B_QUARTER* 22 1089 2.16 1.11 2.00 2.07 1.48 1.00
## DIM_DEHORN 23 596 546.29 339.53 487.00 511.57 316.54 1.00
## INT_DEHORN 24 985 589.93 206.17 587.00 587.22 195.70 0.00
## CALC_PARITY* 25 2043 1.44 0.78 1.00 1.26 0.00 1.00
## C_YEAR* 26 596 6.04 1.57 6.00 6.06 1.48 1.00
## C_MONTH* 27 596 5.38 3.74 4.00 5.11 2.97 1.00
## C_QUARTER* 28 596 1.99 1.04 2.00 1.86 1.48 1.00
## N_DEHORNINGS* 29 2043 1.68 0.82 1.00 1.56 0.00 1.00
## GR_DAYS 30 1582 832.91 602.41 692.00 722.57 269.83 0.00
## G_TOTAL 31 1580 3.09 11.76 2.15 2.13 1.75 0.01
## G_AH_W 32 1576 2.39 8.69 1.71 1.69 1.27 0.01
## G_AH_L 33 1554 0.25 0.58 0.22 0.22 0.07 0.01
## G_AH_C 34 1523 1.05 4.00 0.81 0.82 0.40 0.07
## G_PH_W 35 1385 0.62 2.86 0.32 0.35 0.32 0.01
## G_PH_L 36 1386 0.09 0.31 0.07 0.07 0.03 0.01
## G_PH_C 37 1495 0.88 3.22 0.67 0.69 0.40 0.02
## max range skew kurtosis se
## GENDER* 2.00 1.00 0.56 -1.69 0.01
## AH_W_G 5900.00 5895.00 1.50 4.47 16.35
## PH_W_G 2336.00 2334.00 1.94 7.04 5.77
## SHAVINGS_W_G 495.00 489.00 0.91 2.01 1.33
## TOTAL 8166.00 8161.00 1.29 3.45 21.77
## AH_L 835.00 820.00 2.76 10.17 2.05
## AH_C 870.00 815.00 -0.84 1.98 2.46
## AH_STUMP_L 145.00 135.00 0.47 1.53 0.34
## PH_L 250.00 245.00 2.17 6.48 0.72
## PH_C 960.00 940.00 -0.62 0.46 2.92
## PH_STUMP_L 100.00 95.00 0.97 2.69 0.24
## CALC_AGE 5909.00 5851.00 1.30 2.32 25.45
## CALC_ARRIVAL 5468.00 5786.00 0.80 2.27 15.57
## DH_YEAR* 8.00 7.00 -0.45 -0.89 0.03
## DH_MONTH* 12.00 11.00 -0.30 -1.06 0.07
## DH_QUARTER* 4.00 3.00 -0.27 -1.18 0.02
## W_DIFF 5313.00 5811.00 1.96 7.82 12.05
## L_DIFF 630.00 736.00 2.22 7.45 1.73
## C_DIFF 608.00 1448.00 -1.14 27.34 1.44
## B_YEAR* 16.00 15.00 -0.83 0.78 0.08
## B_MONTH* 12.00 11.00 0.46 -1.01 0.10
## B_QUARTER* 4.00 3.00 0.48 -1.13 0.03
## DIM_DEHORN 2070.00 2069.00 1.05 1.38 13.91
## INT_DEHORN 1335.00 1335.00 0.12 0.74 6.57
## CALC_PARITY* 5.00 4.00 1.79 2.65 0.02
## C_YEAR* 9.00 8.00 -0.13 -0.55 0.06
## C_MONTH* 12.00 11.00 0.60 -1.11 0.15
## C_QUARTER* 4.00 3.00 0.61 -0.91 0.04
## N_DEHORNINGS* 5.00 4.00 1.02 0.42 0.02
## GR_DAYS 4892.00 4892.00 2.98 11.58 15.15
## G_TOTAL 332.67 332.66 24.17 638.75 0.30
## G_AH_W 246.33 246.32 24.25 641.86 0.22
## G_AH_L 18.33 18.32 25.62 727.69 0.01
## G_AH_C 111.67 111.59 26.06 700.90 0.10
## G_PH_W 86.33 86.32 23.35 638.99 0.08
## G_PH_L 10.00 9.99 26.54 788.65 0.01
## G_PH_C 86.67 86.65 25.43 672.70 0.08
summary(df.cv.final[df.cv.final$GENDER == "M", ])## GENDER AH_W_G PH_W_G SHAVINGS_W_G
## F : 0 Min. : 18.0 Min. : 8.0 Min. : 6.0
## M :747 1st Qu.: 853.5 1st Qu.: 101.8 1st Qu.: 55.0
## NA's: 1 Median :1302.0 Median : 263.0 Median :100.0
## Mean :1400.5 Mean : 329.9 Mean :108.8
## 3rd Qu.:1788.5 3rd Qu.: 468.8 3rd Qu.:148.0
## Max. :5830.0 Max. :2336.0 Max. :372.0
## NA's :1 NA's :70 NA's :47
## TOTAL AH_L AH_C AH_STUMP_L
## Min. : 18 Min. : 15 Min. : 80.0 Min. : 35.00
## 1st Qu.:1000 1st Qu.:120 1st Qu.:496.2 1st Qu.: 80.00
## Median :1675 Median :160 Median :575.0 Median : 85.00
## Mean :1802 Mean :174 Mean :580.0 Mean : 89.57
## 3rd Qu.:2361 3rd Qu.:200 3rd Qu.:680.0 3rd Qu.:100.00
## Max. :8166 Max. :600 Max. :870.0 Max. :145.00
## NA's :1 NA's :7 NA's :26 NA's :49
## PH_L PH_C PH_STUMP_L CALC_AGE
## Min. : 10.00 Min. : 20.0 Min. : 5.00 Min. : 58
## 1st Qu.: 38.75 1st Qu.:400.0 1st Qu.: 20.00 1st Qu.: 928
## Median : 50.00 Median :540.0 Median : 30.00 Median :1406
## Mean : 54.53 Mean :519.5 Mean : 28.29 Mean :1576
## 3rd Qu.: 65.00 3rd Qu.:636.2 3rd Qu.: 35.00 3rd Qu.:2053
## Max. :235.00 Max. :800.0 Max. :100.00 Max. :5468
## NA's :72 NA's :36 NA's :105 NA's :160
## CALC_ARRIVAL DH_YEAR DH_MONTH DH_QUARTER W_DIFF
## Min. : 0.0 2014 :206 11 :128 1 :113 Min. :-498.0
## 1st Qu.: 418.8 2015 :196 9 :108 2 :175 1st Qu.: 825.2
## Median : 745.5 2012 :120 6 : 84 3 :193 Median :1083.5
## Mean : 820.7 2013 :112 12 : 84 4 :266 Mean :1158.8
## 3rd Qu.:1169.8 2011 : 84 5 : 68 NA's: 1 3rd Qu.:1400.0
## Max. :5468.0 (Other): 29 (Other):275 Max. :3709.0
## NA's :246 NA's : 1 NA's : 1 NA's :70
## L_DIFF C_DIFF B_YEAR B_MONTH B_QUARTER
## Min. :-65.0 Min. :-140.00 2008 :133 3 : 84 1 :176
## 1st Qu.: 75.0 1st Qu.: 20.00 2011 : 90 4 : 67 2 :177
## Median :110.0 Median : 60.00 2010 : 81 9 : 66 3 :135
## Mean :122.2 Mean : 64.03 2012 : 71 1 : 62 4 :100
## 3rd Qu.:156.2 3rd Qu.: 100.00 2009 : 53 6 : 61 NA's:160
## Max. :418.0 Max. : 409.00 (Other):160 (Other):248
## NA's :72 NA's :37 NA's :160 NA's :160
## DIM_DEHORN INT_DEHORN CALC_PARITY C_YEAR C_MONTH
## Min. : 86 Min. : 35.0 0 :743 2011 : 1 2 : 1
## 1st Qu.:137 1st Qu.: 433.0 1 : 1 2013 : 1 3 : 1
## Median :188 Median : 560.0 2 : 1 2007 : 0 1 : 0
## Mean :188 Mean : 544.2 3 : 0 2008 : 0 4 : 0
## 3rd Qu.:239 3rd Qu.: 653.0 4 : 0 2009 : 0 5 : 0
## Max. :290 Max. :1267.0 NA's: 3 (Other): 0 (Other): 0
## NA's :746 NA's :394 NA's :746 NA's :746
## C_QUARTER N_DEHORNINGS GR_DAYS G_TOTAL
## 1 : 2 1 :391 Min. : 35.0 Min. : 0.2136
## 2 : 0 2 :218 1st Qu.: 546.5 1st Qu.: 0.9866
## 3 : 0 3 :108 Median : 672.0 Median : 2.0951
## 4 : 0 4 : 26 Mean : 849.5 Mean : 2.7944
## NA's:746 5 : 2 3rd Qu.: 932.5 3rd Qu.: 3.7049
## NA's: 3 Max. :4697.0 Max. :39.5676
## NA's :81 NA's :81
## G_AH_W G_AH_L G_AH_C G_PH_W
## Min. : 0.2136 Min. :0.02181 Min. : 0.0779 Min. : 0.01020
## 1st Qu.: 0.8555 1st Qu.:0.19410 1st Qu.: 0.5202 1st Qu.: 0.08737
## Median : 1.6907 Median :0.23309 Median : 0.7977 Median : 0.32456
## Mean : 2.1541 Mean :0.24009 Mean : 0.9128 Mean : 0.53851
## 3rd Qu.: 2.8556 3rd Qu.:0.26881 3rd Qu.: 1.1688 3rd Qu.: 0.69767
## Max. :28.3784 Max. :2.71429 Max. :13.2857 Max. :11.18919
## NA's :81 NA's :86 NA's :101 NA's :147
## G_PH_L G_PH_C
## Min. :0.00860 Min. : 0.01947
## 1st Qu.:0.04320 1st Qu.: 0.41370
## Median :0.06807 Median : 0.68516
## Mean :0.07710 Mean : 0.81241
## 3rd Qu.:0.09511 3rd Qu.: 1.07997
## Max. :1.28571 Max. :13.14286
## NA's :148 NA's :110
describe(df.cv.final[df.cv.final$GENDER == "M", ])## vars n mean sd median trimmed mad min
## GENDER* 1 747 2.00 0.00 2.00 2.00 0.00 2.00
## AH_W_G 2 747 1400.51 776.26 1302.00 1323.66 687.93 18.00
## PH_W_G 3 678 329.86 300.33 263.00 285.08 261.68 8.00
## SHAVINGS_W_G 4 701 108.82 66.53 100.00 102.62 69.68 6.00
## TOTAL 5 747 1802.02 1089.14 1675.00 1689.71 1005.20 18.00
## AH_L 6 741 174.02 75.46 160.00 164.05 59.30 15.00
## AH_C 7 722 580.01 124.90 575.00 582.18 140.85 80.00
## AH_STUMP_L 8 699 89.57 17.11 85.00 88.68 14.83 35.00
## PH_L 9 676 54.53 27.73 50.00 50.59 22.24 10.00
## PH_C 10 712 519.45 144.12 540.00 527.10 163.09 20.00
## PH_STUMP_L 11 643 28.29 11.02 30.00 27.57 14.83 5.00
## CALC_AGE 12 588 1576.42 812.36 1405.50 1494.02 779.85 58.00
## CALC_ARRIVAL 13 502 820.66 627.11 745.50 766.32 571.54 0.00
## DH_YEAR* 14 747 4.37 1.47 5.00 4.48 1.48 1.00
## DH_MONTH* 15 747 7.66 3.30 9.00 7.86 2.97 1.00
## DH_QUARTER* 16 747 2.82 1.08 3.00 2.90 1.48 1.00
## W_DIFF 17 678 1158.75 507.86 1083.50 1105.41 414.39 -498.00
## L_DIFF 18 676 122.24 62.78 110.00 114.97 59.30 -65.00
## C_DIFF 19 711 64.03 63.31 60.00 61.80 59.30 -140.00
## B_YEAR* 20 588 10.31 2.51 11.00 10.46 2.97 2.00
## B_MONTH* 21 588 5.93 3.36 6.00 5.82 4.45 1.00
## B_QUARTER* 22 588 2.27 1.07 2.00 2.21 1.48 1.00
## DIM_DEHORN 23 2 188.00 144.25 188.00 188.00 151.23 86.00
## INT_DEHORN 24 354 544.23 161.20 560.00 546.75 158.64 35.00
## CALC_PARITY* 25 745 1.00 0.08 1.00 1.00 0.00 1.00
## C_YEAR* 26 2 6.00 1.41 6.00 6.00 1.48 5.00
## C_MONTH* 27 2 2.50 0.71 2.50 2.50 0.74 2.00
## C_QUARTER* 28 2 1.00 0.00 1.00 1.00 0.00 1.00
## N_DEHORNINGS* 29 745 1.70 0.86 1.00 1.57 0.00 1.00
## GR_DAYS 30 667 849.48 596.98 672.00 736.92 295.04 35.00
## G_TOTAL 31 667 2.79 3.21 2.10 2.31 1.85 0.21
## G_AH_W 32 667 2.15 2.31 1.69 1.82 1.34 0.21
## G_AH_L 33 662 0.24 0.15 0.23 0.23 0.06 0.02
## G_AH_C 34 647 0.91 0.70 0.80 0.84 0.46 0.08
## G_PH_W 35 601 0.54 0.92 0.32 0.38 0.38 0.01
## G_PH_L 36 600 0.08 0.07 0.07 0.07 0.04 0.01
## G_PH_C 37 638 0.81 0.69 0.69 0.74 0.46 0.02
## max range skew kurtosis se
## GENDER* 2.00 0.00 NaN NaN 0.00
## AH_W_G 5830.00 5812.00 1.41 3.76 28.40
## PH_W_G 2336.00 2328.00 1.99 6.58 11.53
## SHAVINGS_W_G 372.00 366.00 0.89 0.76 2.51
## TOTAL 8166.00 8148.00 1.38 3.55 39.85
## AH_L 600.00 585.00 1.76 5.01 2.77
## AH_C 870.00 790.00 -0.39 0.60 4.65
## AH_STUMP_L 145.00 110.00 0.44 0.20 0.65
## PH_L 235.00 225.00 2.17 7.73 1.07
## PH_C 800.00 780.00 -0.49 -0.38 5.40
## PH_STUMP_L 100.00 95.00 0.99 2.76 0.43
## CALC_AGE 5468.00 5410.00 1.01 1.24 33.50
## CALC_ARRIVAL 5468.00 5468.00 1.58 7.68 27.99
## DH_YEAR* 7.00 6.00 -0.48 -0.83 0.05
## DH_MONTH* 12.00 11.00 -0.41 -0.98 0.12
## DH_QUARTER* 4.00 3.00 -0.36 -1.19 0.04
## W_DIFF 3709.00 4207.00 1.43 3.89 19.50
## L_DIFF 418.00 483.00 1.11 1.54 2.41
## C_DIFF 409.00 549.00 0.60 2.32 2.37
## B_YEAR* 16.00 14.00 -0.51 0.04 0.10
## B_MONTH* 12.00 11.00 0.25 -1.09 0.14
## B_QUARTER* 4.00 3.00 0.29 -1.17 0.04
## DIM_DEHORN 290.00 204.00 0.00 -2.75 102.00
## INT_DEHORN 1267.00 1232.00 -0.05 1.12 8.57
## CALC_PARITY* 3.00 2.00 21.86 500.38 0.00
## C_YEAR* 7.00 2.00 0.00 -2.75 1.00
## C_MONTH* 3.00 1.00 0.00 -2.75 0.50
## C_QUARTER* 1.00 0.00 NaN NaN 0.00
## N_DEHORNINGS* 5.00 4.00 1.05 0.34 0.03
## GR_DAYS 4697.00 4662.00 2.51 7.76 23.12
## G_TOTAL 39.57 39.35 5.68 48.78 0.12
## G_AH_W 28.38 28.16 5.74 49.77 0.09
## G_AH_L 2.71 2.69 9.67 133.84 0.01
## G_AH_C 13.29 13.21 8.87 146.69 0.03
## G_PH_W 11.19 11.18 6.48 56.82 0.04
## G_PH_L 1.29 1.28 9.69 135.22 0.00
## G_PH_C 13.14 13.12 9.31 155.91 0.03
summary(df.cv.final[df.cv.final$GENDER == "F", ])## GENDER AH_W_G PH_W_G SHAVINGS_W_G
## F :1297 Min. : 5 Min. : 2.0 Min. : 10.0
## M : 0 1st Qu.: 845 1st Qu.: 156.0 1st Qu.: 65.0
## NA's: 1 Median :1242 Median : 257.0 Median : 99.0
## Mean :1304 Mean : 296.6 Mean :100.9
## 3rd Qu.:1593 3rd Qu.: 386.5 3rd Qu.:133.0
## Max. :5900 Max. :1348.0 Max. :495.0
## NA's :5 NA's :147 NA's :129
## TOTAL AH_L AH_C AH_STUMP_L
## Min. : 5 Min. : 20.0 Min. : 55.0 Min. : 10.00
## 1st Qu.:1038 1st Qu.:110.0 1st Qu.:515.0 1st Qu.: 75.00
## Median :1594 Median :140.0 Median :600.0 Median : 80.00
## Mean :1655 Mean :165.8 Mean :568.4 Mean : 81.84
## 3rd Qu.:2089 3rd Qu.:175.0 3rd Qu.:635.0 3rd Qu.: 90.00
## Max. :6696 Max. :835.0 Max. :780.0 Max. :130.00
## NA's :1 NA's :31 NA's :65 NA's :137
## PH_L PH_C PH_STUMP_L CALC_AGE
## Min. : 5.00 Min. : 30.0 Min. :10.00 Min. : 94
## 1st Qu.: 40.00 1st Qu.:423.8 1st Qu.:20.00 1st Qu.: 896
## Median : 50.00 Median :500.0 Median :25.00 Median :1349
## Mean : 57.82 Mean :474.5 Mean :26.62 Mean :1529
## 3rd Qu.: 65.00 3rd Qu.:550.0 3rd Qu.:30.00 3rd Qu.:1880
## Max. :250.00 Max. :960.0 Max. :80.00 Max. :5909
## NA's :151 NA's :94 NA's :239 NA's :797
## CALC_ARRIVAL DH_YEAR DH_MONTH DH_QUARTER W_DIFF
## Min. :-318.0 2014 :341 9 :201 1 :170 Min. : 83.0
## 1st Qu.: 517.0 2015 :335 11 :188 2 :377 1st Qu.: 779.5
## Median : 824.0 2013 :228 5 :183 3 :349 Median :1023.0
## Mean : 911.8 2012 :199 6 :111 4 :401 Mean :1106.6
## 3rd Qu.:1331.0 2011 :168 12 :108 NA's: 1 3rd Qu.:1301.0
## Max. :2636.0 (Other): 26 (Other):506 Max. :5313.0
## NA's :253 NA's : 1 NA's : 1 NA's :151
## L_DIFF C_DIFF B_YEAR B_MONTH B_QUARTER
## Min. :-106.0 Min. :-840.00 2010 :120 2 : 85 1 :224
## 1st Qu.: 65.0 1st Qu.: 60.00 2011 :100 3 : 85 2 :136
## Median : 85.0 Median : 95.00 2012 : 66 4 : 57 3 : 48
## Mean : 108.2 Mean : 98.12 2008 : 58 6 : 55 4 : 93
## 3rd Qu.: 120.0 3rd Qu.: 130.00 2009 : 45 1 : 54 NA's:797
## Max. : 630.0 Max. : 608.00 (Other):112 (Other):165
## NA's :155 NA's :98 NA's :797 NA's :797
## DIM_DEHORN INT_DEHORN CALC_PARITY C_YEAR C_MONTH
## Min. : 1.0 Min. : 0.0 0 :703 2011 :131 2 : 93
## 1st Qu.: 286.0 1st Qu.: 467.0 1 :355 2012 :122 3 : 92
## Median : 487.0 Median : 616.0 2 :186 2013 :113 4 : 82
## Mean : 547.5 Mean : 615.6 3 : 47 2014 : 99 1 : 71
## 3rd Qu.: 722.5 3rd Qu.: 754.0 4 : 6 2010 : 79 12 : 66
## Max. :2070.0 Max. :1335.0 NA's: 1 (Other): 50 (Other):190
## NA's :704 NA's :667 NA's :704 NA's :704
## C_QUARTER N_DEHORNINGS GR_DAYS G_TOTAL
## 1 :256 1 :666 Min. : 0.0 Min. : 0.0090
## 2 :152 2 :430 1st Qu.: 539.0 1st Qu.: 0.8115
## 3 :120 3 :176 Median : 702.0 Median : 2.1719
## 4 : 66 4 : 22 Mean : 820.8 Mean : 3.3066
## NA's:704 5 : 3 3rd Qu.: 888.5 3rd Qu.: 2.8868
## NA's: 1 Max. :4892.0 Max. :332.6667
## NA's :383 NA's :385
## G_AH_W G_AH_L G_AH_C G_PH_W
## Min. : 0.009 Min. : 0.0145 Min. : 0.0731 Min. : 0.0098
## 1st Qu.: 0.743 1st Qu.: 0.1623 1st Qu.: 0.5595 1st Qu.: 0.1364
## Median : 1.713 Median : 0.2004 Median : 0.8206 Median : 0.3166
## Mean : 2.570 Mean : 0.2601 Mean : 1.1430 Mean : 0.6847
## 3rd Qu.: 2.272 3rd Qu.: 0.2493 3rd Qu.: 1.0250 3rd Qu.: 0.4729
## Max. :246.333 Max. :18.3333 Max. :111.6667 Max. :86.3333
## NA's :389 NA's :406 NA's :422 NA's :514
## G_PH_L G_PH_C
## Min. : 0.0073 Min. : 0.0274
## 1st Qu.: 0.0472 1st Qu.: 0.3927
## Median : 0.0660 Median : 0.6723
## Mean : 0.0964 Mean : 0.9298
## 3rd Qu.: 0.0849 3rd Qu.: 0.8602
## Max. :10.0000 Max. :86.6667
## NA's :512 NA's :441
describe(df.cv.final[df.cv.final$GENDER == "F", ])## vars n mean sd median trimmed mad min
## GENDER* 1 1297 1.00 0.00 1.00 1.00 0.00 1.00
## AH_W_G 2 1293 1304.37 713.42 1242.00 1236.49 551.53 5.00
## PH_W_G 3 1151 296.61 208.19 257.00 271.27 163.09 2.00
## SHAVINGS_W_G 4 1169 100.93 51.38 99.00 98.91 50.41 10.00
## TOTAL 5 1297 1654.54 914.23 1594.00 1580.59 779.85 5.00
## AH_L 6 1267 165.77 99.80 140.00 144.77 44.48 20.00
## AH_C 7 1233 568.38 97.85 600.00 579.56 74.13 55.00
## AH_STUMP_L 8 1161 81.84 11.90 80.00 81.87 14.83 10.00
## PH_L 9 1147 57.82 32.40 50.00 52.25 14.83 5.00
## PH_C 10 1204 474.55 113.84 500.00 487.80 88.96 30.00
## PH_STUMP_L 11 1059 26.62 9.44 25.00 26.17 7.41 10.00
## CALC_AGE 12 501 1529.31 871.36 1349.00 1404.41 722.03 94.00
## CALC_ARRIVAL 13 1045 911.79 603.16 824.00 881.29 598.97 -318.00
## DH_YEAR* 14 1297 4.32 1.43 5.00 4.42 1.48 1.00
## DH_MONTH* 15 1297 7.39 3.17 7.00 7.52 4.45 1.00
## DH_QUARTER* 16 1297 2.76 1.03 3.00 2.82 1.48 1.00
## W_DIFF 17 1147 1106.60 518.11 1023.00 1044.93 382.51 83.00
## L_DIFF 18 1143 108.24 79.19 85.00 93.02 37.06 -106.00
## C_DIFF 19 1200 98.12 59.12 95.00 96.78 51.89 -840.00
## B_YEAR* 20 501 10.83 2.45 11.00 11.10 1.48 1.00
## B_MONTH* 21 501 5.16 3.52 4.00 4.83 2.97 1.00
## B_QUARTER* 22 501 2.02 1.13 2.00 1.90 1.48 1.00
## DIM_DEHORN 23 594 547.50 339.41 487.00 512.77 316.54 1.00
## INT_DEHORN 24 631 615.57 223.60 616.00 614.64 212.01 0.00
## CALC_PARITY* 25 1297 1.69 0.88 1.00 1.55 0.00 1.00
## C_YEAR* 26 594 6.04 1.57 6.00 6.06 1.48 1.00
## C_MONTH* 27 594 5.39 3.74 4.00 5.12 2.97 1.00
## C_QUARTER* 28 594 1.99 1.04 2.00 1.87 1.48 1.00
## N_DEHORNINGS* 29 1297 1.66 0.79 1.00 1.55 0.00 1.00
## GR_DAYS 30 915 820.83 606.38 702.00 714.67 259.45 0.00
## G_TOTAL 31 913 3.31 15.23 2.17 2.02 1.52 0.01
## G_AH_W 32 909 2.57 11.27 1.71 1.62 1.09 0.01
## G_AH_L 33 892 0.26 0.76 0.20 0.21 0.06 0.01
## G_AH_C 34 876 1.14 5.24 0.82 0.81 0.34 0.07
## G_PH_W 35 784 0.68 3.72 0.32 0.32 0.24 0.01
## G_PH_L 36 786 0.10 0.41 0.07 0.07 0.03 0.01
## G_PH_C 37 857 0.93 4.21 0.67 0.66 0.34 0.03
## max range skew kurtosis se
## GENDER* 1.00 0.00 NaN NaN 0.00
## AH_W_G 5900.00 5895.00 1.55 4.96 19.84
## PH_W_G 1348.00 1346.00 1.42 2.92 6.14
## SHAVINGS_W_G 495.00 485.00 0.77 3.00 1.50
## TOTAL 6696.00 6691.00 1.12 2.69 25.39
## AH_L 835.00 815.00 2.99 10.57 2.80
## AH_C 780.00 725.00 -1.45 3.29 2.79
## AH_STUMP_L 130.00 120.00 -0.21 2.00 0.35
## PH_L 250.00 245.00 2.13 5.75 0.96
## PH_C 960.00 930.00 -1.07 1.22 3.28
## PH_STUMP_L 80.00 70.00 0.87 2.19 0.29
## CALC_AGE 5909.00 5815.00 1.58 3.30 38.93
## CALC_ARRIVAL 2636.00 2954.00 0.40 -0.49 18.66
## DH_YEAR* 8.00 7.00 -0.43 -0.93 0.04
## DH_MONTH* 12.00 11.00 -0.24 -1.09 0.09
## DH_QUARTER* 4.00 3.00 -0.22 -1.17 0.03
## W_DIFF 5313.00 5230.00 2.27 10.13 15.30
## L_DIFF 630.00 736.00 2.61 8.98 2.34
## C_DIFF 608.00 1448.00 -2.51 56.71 1.71
## B_YEAR* 15.00 14.00 -1.25 2.09 0.11
## B_MONTH* 12.00 11.00 0.72 -0.77 0.16
## B_QUARTER* 4.00 3.00 0.72 -0.93 0.05
## DIM_DEHORN 2070.00 2069.00 1.05 1.38 13.93
## INT_DEHORN 1335.00 1335.00 -0.01 0.44 8.90
## CALC_PARITY* 5.00 4.00 1.13 0.56 0.02
## C_YEAR* 9.00 8.00 -0.13 -0.55 0.06
## C_MONTH* 12.00 11.00 0.59 -1.12 0.15
## C_QUARTER* 4.00 3.00 0.61 -0.91 0.04
## N_DEHORNINGS* 5.00 4.00 0.99 0.41 0.02
## GR_DAYS 4892.00 4892.00 3.31 14.26 20.05
## G_TOTAL 332.67 332.66 19.20 391.14 0.50
## G_AH_W 246.33 246.32 19.20 391.03 0.37
## G_AH_L 18.33 18.32 20.22 441.43 0.03
## G_AH_C 111.67 111.59 20.10 411.45 0.18
## G_PH_W 86.33 86.32 18.69 393.65 0.13
## G_PH_L 10.00 9.99 20.66 467.61 0.01
## G_PH_C 86.67 86.64 19.77 398.97 0.14
summary(df.cv.final[df.cv.final$N_DEHORNINGS == "1", ])## GENDER AH_W_G PH_W_G SHAVINGS_W_G
## F :666 Min. : 12.0 Min. : 2.0 Min. : 6.00
## M :391 1st Qu.: 667.2 1st Qu.: 69.0 1st Qu.: 40.00
## NA's: 3 Median :1075.5 Median : 246.0 Median : 75.00
## Mean :1243.8 Mean : 287.7 Mean : 85.23
## 3rd Qu.:1578.2 3rd Qu.: 408.0 3rd Qu.:118.00
## Max. :5900.0 Max. :2336.0 Max. :495.00
## NA's :4 NA's :191 NA's :82
## TOTAL AH_L AH_C AH_STUMP_L
## Min. : 12 Min. : 15.0 Min. : 55.0 Min. : 10.00
## 1st Qu.: 740 1st Qu.:135.0 1st Qu.:470.0 1st Qu.: 75.00
## Median :1330 Median :170.0 Median :550.0 Median : 80.00
## Mean :1557 Mean :202.6 Mean :544.5 Mean : 80.83
## 3rd Qu.:2067 3rd Qu.:230.0 3rd Qu.:630.0 3rd Qu.: 90.00
## Max. :8166 Max. :835.0 Max. :870.0 Max. :140.00
## NA's :2 NA's :20 NA's :59 NA's :91
## PH_L PH_C PH_STUMP_L CALC_AGE
## Min. : 5.00 Min. : 20.0 Min. : 5.00 Min. : 58
## 1st Qu.: 40.00 1st Qu.:360.0 1st Qu.: 20.00 1st Qu.: 760
## Median : 55.00 Median :495.0 Median : 25.00 Median : 966
## Mean : 65.54 Mean :461.9 Mean : 25.51 Mean :1234
## 3rd Qu.: 80.00 3rd Qu.:560.0 3rd Qu.: 30.00 3rd Qu.:1415
## Max. :250.00 Max. :960.0 Max. :100.00 Max. :4892
## NA's :198 NA's :98 NA's :243 NA's :463
## CALC_ARRIVAL DH_YEAR DH_MONTH DH_QUARTER W_DIFF
## Min. :-318.00 2014 :250 11 :211 1 :141 Min. :-498
## 1st Qu.: 66.75 2011 :236 9 :164 2 :236 1st Qu.: 722
## Median : 520.00 2012 :197 5 :116 3 :290 Median : 954
## Mean : 489.74 2015 :168 10 : 93 4 :391 Mean :1107
## 3rd Qu.: 760.50 2013 :162 12 : 87 NA's: 2 3rd Qu.:1324
## Max. :4697.00 (Other): 45 (Other):387 Max. :5313
## NA's :314 NA's : 2 NA's : 2 NA's :193
## L_DIFF C_DIFF B_YEAR B_MONTH B_QUARTER
## Min. :-106.0 Min. :-840.00 2011 :109 3 : 90 1 :228
## 1st Qu.: 80.0 1st Qu.: 50.00 2012 :102 4 : 70 2 :159
## Median : 130.0 Median : 90.00 2013 : 88 1 : 69 3 : 91
## Mean : 145.9 Mean : 89.59 2010 : 86 2 : 69 4 :119
## 3rd Qu.: 190.0 3rd Qu.: 127.00 2008 : 66 12 : 56 NA's:463
## Max. : 630.0 Max. : 300.00 (Other):146 (Other):243
## NA's :200 NA's :101 NA's :463 NA's :463
## DIM_DEHORN INT_DEHORN CALC_PARITY C_YEAR C_MONTH
## Min. : 3.0 Min. : NA 0 :831 2011 : 72 3 : 34
## 1st Qu.: 336.0 1st Qu.: NA 1 :174 2010 : 63 2 : 31
## Median : 518.0 Median : NA 2 : 48 2012 : 31 4 : 29
## Mean : 571.6 Mean :NaN 3 : 4 2013 : 21 1 : 22
## 3rd Qu.: 731.0 3rd Qu.: NA 4 : 1 2014 : 20 12 : 22
## Max. :1785.0 Max. : NA NA's: 2 (Other): 20 (Other): 89
## NA's :833 NA's :1060 NA's :833 NA's :833
## C_QUARTER N_DEHORNINGS GR_DAYS G_TOTAL G_AH_W
## 1 : 87 1 :1058 Min. : 58 Min. :0.0119 Min. :0.1148
## 2 : 69 2 : 0 1st Qu.: 760 1st Qu.:0.5909 1st Qu.:0.5255
## 3 : 49 3 : 0 Median : 966 Median :0.7870 Median :0.6866
## 4 : 22 4 : 0 Mean :1234 Mean :0.8446 Mean :0.7390
## NA's:833 5 : 0 3rd Qu.:1415 3rd Qu.:1.0322 3rd Qu.:0.9116
## NA's: 2 Max. :4892 Max. :3.9301 Max. :2.8694
## NA's :463 NA's :463 NA's :465
## G_AH_L G_AH_C G_PH_W G_PH_L
## Min. :0.0145 Min. :0.0731 Min. :0.0098 Min. :0.0073
## 1st Qu.:0.1476 1st Qu.:0.3517 1st Qu.:0.0364 1st Qu.:0.0257
## Median :0.2020 Median :0.4867 Median :0.0648 Median :0.0359
## Mean :0.1920 Mean :0.4823 Mean :0.0927 Mean :0.0390
## 3rd Qu.:0.2369 3rd Qu.:0.5797 3rd Qu.:0.1154 3rd Qu.:0.0488
## Max. :0.5714 Max. :1.6489 Max. :0.9409 Max. :0.1117
## NA's :470 NA's :487 NA's :631 NA's :631
## G_PH_C
## Min. :0.0195
## 1st Qu.:0.2676
## Median :0.3463
## Mean :0.3578
## 3rd Qu.:0.4440
## Max. :1.2381
## NA's :515
describe(df.cv.final[df.cv.final$N_DEHORNINGS == "1", ])## Warning in FUN(newX[, i], ...): no non-missing arguments to min; returning
## Inf
## Warning in FUN(newX[, i], ...): no non-missing arguments to max; returning
## -Inf
## vars n mean sd median trimmed mad min
## GENDER* 1 1057 1.37 0.48 1.00 1.34 0.00 1.00
## AH_W_G 2 1056 1243.84 849.85 1075.50 1123.93 655.31 12.00
## PH_W_G 3 869 287.70 261.93 246.00 249.70 255.01 2.00
## SHAVINGS_W_G 4 978 85.23 58.54 75.00 78.42 57.82 6.00
## TOTAL 5 1058 1556.58 1110.01 1330.50 1405.26 962.21 12.00
## AH_L 6 1040 202.61 112.72 170.00 182.58 59.30 15.00
## AH_C 7 1001 544.46 117.93 550.00 550.06 118.61 55.00
## AH_STUMP_L 8 969 80.83 12.49 80.00 80.69 7.41 10.00
## PH_L 9 862 65.54 39.50 55.00 59.29 29.65 5.00
## PH_C 10 962 461.87 138.34 495.00 467.58 133.43 20.00
## PH_STUMP_L 11 817 25.51 9.50 25.00 24.75 7.41 5.00
## CALC_AGE 12 597 1233.80 796.19 966.00 1089.28 378.06 58.00
## CALC_ARRIVAL 13 746 489.74 419.62 520.00 455.92 487.78 -318.00
## DH_YEAR* 14 1058 3.80 1.52 4.00 3.80 1.48 1.00
## DH_MONTH* 15 1058 7.73 3.24 9.00 7.98 2.97 1.00
## DH_QUARTER* 16 1058 2.88 1.05 3.00 2.97 1.48 1.00
## W_DIFF 17 867 1107.20 623.75 954.00 1016.94 415.13 -498.00
## L_DIFF 18 860 145.95 92.60 130.00 134.63 74.13 -106.00
## C_DIFF 19 959 89.59 67.46 90.00 88.88 59.30 -840.00
## B_YEAR* 20 597 11.09 2.60 12.00 11.39 2.97 1.00
## B_MONTH* 21 597 5.58 3.57 5.00 5.37 4.45 1.00
## B_QUARTER* 22 597 2.17 1.14 2.00 2.09 1.48 1.00
## DIM_DEHORN 23 227 571.60 327.64 518.00 541.40 292.07 3.00
## INT_DEHORN 24 0 NaN NA NA NaN NA Inf
## CALC_PARITY* 25 1058 1.27 0.57 1.00 1.14 0.00 1.00
## C_YEAR* 26 227 5.12 1.48 5.00 5.06 1.48 1.00
## C_MONTH* 27 227 5.53 3.55 4.00 5.30 2.97 1.00
## C_QUARTER* 28 227 2.03 1.00 2.00 1.91 1.48 1.00
## N_DEHORNINGS* 29 1058 1.00 0.00 1.00 1.00 0.00 1.00
## GR_DAYS 30 597 1233.80 796.19 966.00 1089.28 378.06 58.00
## G_TOTAL 31 597 0.84 0.38 0.79 0.81 0.32 0.01
## G_AH_W 32 595 0.74 0.31 0.69 0.71 0.28 0.11
## G_AH_L 33 590 0.19 0.08 0.20 0.19 0.06 0.01
## G_AH_C 34 573 0.48 0.20 0.49 0.47 0.17 0.07
## G_PH_W 35 429 0.09 0.10 0.06 0.08 0.05 0.01
## G_PH_L 36 429 0.04 0.02 0.04 0.04 0.02 0.01
## G_PH_C 37 545 0.36 0.13 0.35 0.35 0.13 0.02
## max range skew kurtosis se
## GENDER* 2.00 1.00 0.54 -1.71 0.01
## AH_W_G 5900.00 5888.00 1.75 4.47 26.15
## PH_W_G 2336.00 2334.00 1.83 6.40 8.89
## SHAVINGS_W_G 495.00 489.00 1.48 4.23 1.87
## TOTAL 8166.00 8154.00 1.55 3.57 34.13
## AH_L 835.00 820.00 2.04 5.04 3.50
## AH_C 870.00 815.00 -0.75 1.74 3.73
## AH_STUMP_L 140.00 130.00 0.22 3.57 0.40
## PH_L 250.00 245.00 1.54 2.53 1.35
## PH_C 960.00 940.00 -0.42 -0.07 4.46
## PH_STUMP_L 100.00 95.00 1.43 5.65 0.33
## CALC_AGE 4892.00 4834.00 1.93 4.09 32.59
## CALC_ARRIVAL 4697.00 5015.00 1.76 12.91 15.36
## DH_YEAR* 8.00 7.00 -0.04 -1.21 0.05
## DH_MONTH* 12.00 11.00 -0.49 -0.94 0.10
## DH_QUARTER* 4.00 3.00 -0.44 -1.08 0.03
## W_DIFF 5313.00 5811.00 2.07 6.78 21.18
## L_DIFF 630.00 736.00 1.40 3.11 3.16
## C_DIFF 300.00 1140.00 -2.55 36.96 2.18
## B_YEAR* 16.00 15.00 -1.02 0.93 0.11
## B_MONTH* 12.00 11.00 0.46 -1.11 0.15
## B_QUARTER* 4.00 3.00 0.47 -1.22 0.05
## DIM_DEHORN 1785.00 1782.00 0.88 0.71 21.75
## INT_DEHORN -Inf -Inf NA NA NA
## CALC_PARITY* 5.00 4.00 2.24 5.18 0.02
## C_YEAR* 9.00 8.00 0.30 -0.09 0.10
## C_MONTH* 12.00 11.00 0.53 -1.06 0.24
## C_QUARTER* 4.00 3.00 0.54 -0.87 0.07
## N_DEHORNINGS* 1.00 0.00 NaN NaN 0.00
## GR_DAYS 4892.00 4834.00 1.93 4.09 32.59
## G_TOTAL 3.93 3.92 1.96 10.01 0.02
## G_AH_W 2.87 2.75 1.42 5.17 0.01
## G_AH_L 0.57 0.56 0.22 0.96 0.00
## G_AH_C 1.65 1.58 1.29 4.98 0.01
## G_PH_W 0.94 0.93 3.93 24.11 0.00
## G_PH_L 0.11 0.10 1.01 1.28 0.00
## G_PH_C 1.24 1.22 0.78 3.26 0.01
summary(df.cv.final[df.cv.final$N_DEHORNINGS == "2", ])## GENDER AH_W_G PH_W_G SHAVINGS_W_G
## F :430 Min. : 12 Min. : 23.0 Min. : 11.00
## M :218 1st Qu.: 963 1st Qu.: 157.5 1st Qu.: 80.75
## NA's: 2 Median :1303 Median : 251.0 Median :108.50
## Mean :1355 Mean : 294.6 Mean :116.50
## 3rd Qu.:1629 3rd Qu.: 365.0 3rd Qu.:142.00
## Max. :4412 Max. :1359.0 Max. :347.00
## NA's :4 NA's :23 NA's :58
## TOTAL AH_L AH_C AH_STUMP_L
## Min. : 12 Min. : 27.0 Min. :160 Min. : 40.00
## 1st Qu.:1204 1st Qu.:105.0 1st Qu.:545 1st Qu.: 80.00
## Median :1656 Median :125.0 Median :600 Median : 85.00
## Mean :1742 Mean :129.2 Mean :593 Mean : 86.61
## 3rd Qu.:2132 3rd Qu.:150.0 3rd Qu.:650 3rd Qu.: 95.00
## Max. :5494 Max. :280.0 Max. :840 Max. :140.00
## NA's :2 NA's :15 NA's :25 NA's :59
## PH_L PH_C PH_STUMP_L CALC_AGE
## Min. : 15.00 Min. : 52.0 Min. :10.00 Min. : 778
## 1st Qu.: 35.00 1st Qu.:460.0 1st Qu.:20.00 1st Qu.:1340
## Median : 45.00 Median :520.0 Median :30.00 Median :1554
## Mean : 46.45 Mean :509.7 Mean :28.13 Mean :1767
## 3rd Qu.: 55.00 3rd Qu.:580.0 3rd Qu.:35.00 3rd Qu.:1868
## Max. :160.00 Max. :780.0 Max. :80.00 Max. :5909
## NA's :21 NA's :24 NA's :65 NA's :334
## CALC_ARRIVAL DH_YEAR DH_MONTH DH_QUARTER W_DIFF
## Min. : 16 2015 :187 9 : 92 1 : 84 Min. : 348
## 1st Qu.: 706 2014 :180 11 : 83 2 :196 1st Qu.: 811
## Median :1022 2013 :146 5 : 82 3 :167 Median :1058
## Mean :1073 2012 :111 6 : 73 4 :201 Mean :1092
## 3rd Qu.:1358 2011 : 16 12 : 68 NA's: 2 3rd Qu.:1285
## Max. :5468 (Other): 8 (Other):250 Max. :3330
## NA's :133 NA's : 2 NA's : 2 NA's :25
## L_DIFF C_DIFF B_YEAR B_MONTH B_QUARTER
## Min. : -3.00 Min. :-95.00 2011 : 73 3 : 48 1 :111
## 1st Qu.: 60.00 1st Qu.: 45.00 2010 : 68 4 : 37 2 : 93
## Median : 80.00 Median : 80.00 2008 : 60 6 : 36 3 : 58
## Mean : 83.17 Mean : 82.79 2012 : 35 2 : 33 4 : 54
## 3rd Qu.:100.00 3rd Qu.:120.00 2009 : 29 1 : 30 NA's:334
## Max. :230.00 Max. :608.00 (Other): 51 (Other):132
## NA's :23 NA's :26 NA's :334 NA's :334
## DIM_DEHORN INT_DEHORN CALC_PARITY C_YEAR C_MONTH
## Min. : 7.0 Min. : 0.0 0 :405 2012 : 74 2 : 38
## 1st Qu.: 280.0 1st Qu.: 448.0 1 :132 2013 : 57 3 : 35
## Median : 503.0 Median : 580.0 2 : 81 2011 : 53 1 : 33
## Mean : 549.6 Mean : 594.4 3 : 28 2014 : 34 4 : 33
## 3rd Qu.: 724.0 3rd Qu.: 720.2 4 : 2 2010 : 14 12 : 30
## Max. :1770.0 Max. :1335.0 NA's: 2 (Other): 11 (Other): 74
## NA's :407 NA's :2 NA's :407 NA's :407
## C_QUARTER N_DEHORNINGS GR_DAYS G_TOTAL
## 1 :106 1 : 0 Min. : 0.0 Min. : 0.0511
## 2 : 56 2 :648 1st Qu.: 448.0 1st Qu.: 2.1389
## 3 : 51 3 : 0 Median : 580.0 Median : 2.6757
## 4 : 30 4 : 0 Mean : 594.4 Mean : 3.9411
## NA's:407 5 : 0 3rd Qu.: 720.2 3rd Qu.: 3.5625
## NA's: 2 Max. :1335.0 Max. :332.6667
## NA's :2 NA's :3
## G_AH_W G_AH_L G_AH_C
## Min. : 0.05112 Min. : 0.06522 Min. : 0.1573
## 1st Qu.: 1.70190 1st Qu.: 0.17857 1st Qu.: 0.8107
## Median : 2.12825 Median : 0.22412 Median : 0.9770
## Mean : 3.03456 Mean : 0.26914 Mean : 1.3018
## 3rd Qu.: 2.78798 3rd Qu.: 0.27356 3rd Qu.: 1.2758
## Max. :246.33333 Max. :12.50000 Max. :111.6667
## NA's :5 NA's :16 NA's :26
## G_PH_W G_PH_L G_PH_C
## Min. : 0.03932 Min. :0.02408 Min. : 0.07439
## 1st Qu.: 0.26451 1st Qu.:0.06074 1st Qu.: 0.67227
## Median : 0.39969 Median :0.07435 Median : 0.83449
## Mean : 0.75066 Mean :0.09806 Mean : 1.10213
## 3rd Qu.: 0.61382 3rd Qu.:0.09440 3rd Qu.: 1.10512
## Max. :86.33333 Max. :5.00000 Max. :86.66667
## NA's :24 NA's :22 NA's :25
describe(df.cv.final[df.cv.final$N_DEHORNINGS == "2", ])## vars n mean sd median trimmed mad min
## GENDER* 1 648 1.34 0.47 1.00 1.30 0.00 1.00
## AH_W_G 2 646 1354.83 559.81 1303.00 1309.95 492.22 12.00
## PH_W_G 3 627 294.59 204.24 251.00 265.17 149.74 23.00
## SHAVINGS_W_G 4 592 116.50 48.94 108.50 112.32 45.22 11.00
## TOTAL 5 648 1742.13 760.46 1655.50 1683.42 690.15 12.00
## AH_L 6 635 129.15 34.00 125.00 126.89 37.06 27.00
## AH_C 7 625 592.98 89.90 600.00 597.25 74.13 160.00
## AH_STUMP_L 8 591 86.61 14.30 85.00 86.05 14.83 40.00
## PH_L 9 629 46.45 15.67 45.00 45.29 14.83 15.00
## PH_C 10 626 509.67 106.82 520.00 516.64 88.96 52.00
## PH_STUMP_L 11 585 28.13 10.22 30.00 27.70 14.83 10.00
## CALC_AGE 12 316 1766.76 749.66 1554.50 1626.83 341.00 778.00
## CALC_ARRIVAL 13 517 1072.92 497.79 1022.00 1037.38 492.22 16.00
## DH_YEAR* 14 648 4.67 1.17 5.00 4.73 1.48 2.00
## DH_MONTH* 15 648 7.39 3.19 7.00 7.50 4.45 1.00
## DH_QUARTER* 16 648 2.75 1.03 3.00 2.81 1.48 1.00
## W_DIFF 17 625 1092.21 388.43 1058.00 1056.43 352.86 348.00
## L_DIFF 18 627 83.17 29.10 80.00 81.65 29.65 -3.00
## C_DIFF 19 624 82.79 57.10 80.00 82.57 59.30 -95.00
## B_YEAR* 20 316 10.22 2.34 11.00 10.52 1.48 1.00
## B_MONTH* 21 316 5.64 3.39 5.00 5.45 4.45 1.00
## B_QUARTER* 22 316 2.17 1.09 2.00 2.09 1.48 1.00
## DIM_DEHORN 23 243 549.62 336.04 503.00 518.67 329.14 7.00
## INT_DEHORN 24 648 594.39 214.89 580.00 586.05 201.63 0.00
## CALC_PARITY* 25 648 1.60 0.89 1.00 1.43 0.00 1.00
## C_YEAR* 26 243 6.27 1.26 6.00 6.26 1.48 3.00
## C_MONTH* 27 243 5.50 3.91 4.00 5.26 2.97 1.00
## C_QUARTER* 28 243 2.02 1.07 2.00 1.90 1.48 1.00
## N_DEHORNINGS* 29 648 2.00 0.00 2.00 2.00 0.00 2.00
## GR_DAYS 30 648 594.39 214.89 580.00 586.05 201.63 0.00
## G_TOTAL 31 647 3.94 13.39 2.68 2.86 1.01 0.05
## G_AH_W 32 645 3.03 9.91 2.13 2.24 0.75 0.05
## G_AH_L 33 634 0.27 0.53 0.22 0.22 0.07 0.07
## G_AH_C 34 624 1.30 4.49 0.98 1.03 0.31 0.16
## G_PH_W 35 626 0.75 3.56 0.40 0.45 0.24 0.04
## G_PH_L 36 628 0.10 0.22 0.07 0.08 0.02 0.02
## G_PH_C 37 625 1.10 3.49 0.83 0.88 0.30 0.07
## max range skew kurtosis se
## GENDER* 2.00 1.00 0.69 -1.53 0.02
## AH_W_G 4412.00 4400.00 1.10 3.06 22.03
## PH_W_G 1359.00 1336.00 1.70 3.86 8.16
## SHAVINGS_W_G 347.00 336.00 1.05 1.99 2.01
## TOTAL 5494.00 5482.00 1.00 2.32 29.87
## AH_L 280.00 253.00 0.73 0.95 1.35
## AH_C 840.00 680.00 -0.54 1.18 3.60
## AH_STUMP_L 140.00 100.00 0.33 0.82 0.59
## PH_L 160.00 145.00 1.16 4.33 0.62
## PH_C 780.00 728.00 -0.70 1.13 4.27
## PH_STUMP_L 80.00 70.00 0.69 1.23 0.42
## CALC_AGE 5909.00 5131.00 2.42 7.10 42.17
## CALC_ARRIVAL 5468.00 5452.00 1.68 10.67 21.89
## DH_YEAR* 7.00 5.00 -0.31 -0.90 0.05
## DH_MONTH* 12.00 11.00 -0.19 -1.08 0.13
## DH_QUARTER* 4.00 3.00 -0.19 -1.19 0.04
## W_DIFF 3330.00 2982.00 1.28 3.39 15.54
## L_DIFF 230.00 233.00 0.57 0.92 1.16
## C_DIFF 608.00 703.00 1.42 12.03 2.29
## B_YEAR* 14.00 13.00 -1.25 1.64 0.13
## B_MONTH* 12.00 11.00 0.42 -1.03 0.19
## B_QUARTER* 4.00 3.00 0.44 -1.13 0.06
## DIM_DEHORN 1770.00 1763.00 0.92 0.92 21.56
## INT_DEHORN 1335.00 1335.00 0.32 0.62 8.44
## CALC_PARITY* 5.00 4.00 1.36 0.91 0.03
## C_YEAR* 9.00 6.00 0.08 -0.45 0.08
## C_MONTH* 12.00 11.00 0.52 -1.30 0.25
## C_QUARTER* 4.00 3.00 0.56 -1.04 0.07
## N_DEHORNINGS* 2.00 0.00 NaN NaN 0.00
## GR_DAYS 1335.00 1335.00 0.32 0.62 8.44
## G_TOTAL 332.67 332.62 22.96 558.16 0.53
## G_AH_W 246.33 246.28 23.03 560.33 0.39
## G_AH_L 12.50 12.43 20.07 452.66 0.02
## G_AH_C 111.67 111.51 23.88 583.41 0.18
## G_PH_W 86.33 86.29 22.39 533.20 0.14
## G_PH_L 5.00 4.98 18.19 373.38 0.01
## G_PH_C 86.67 86.59 23.55 572.05 0.14
summary(df.cv.final[!df.cv.final$N_DEHORNINGS %in% c("1", "2"),])## GENDER AH_W_G PH_W_G SHAVINGS_W_G TOTAL
## F:201 Min. : 5 Min. : 47.0 Min. : 22.0 Min. : 5
## M:138 1st Qu.:1262 1st Qu.: 226.0 1st Qu.:110.0 1st Qu.:1627
## Median :1520 Median : 315.0 Median :133.0 Median :1982
## Mean :1605 Mean : 391.4 Mean :139.8 Mean :2113
## 3rd Qu.:1892 3rd Qu.: 494.0 3rd Qu.:167.2 3rd Qu.:2480
## Max. :4654 Max. :1979.0 Max. :322.0 Max. :6633
## NA's :6 NA's :39
## AH_L AH_C AH_STUMP_L PH_L
## Min. : 60.0 Min. :340.0 Min. : 55.00 Min. : 10.00
## 1st Qu.:110.0 1st Qu.:585.0 1st Qu.: 80.00 1st Qu.: 40.00
## Median :135.0 Median :630.0 Median : 90.00 Median : 50.00
## Mean :138.9 Mean :619.9 Mean : 93.72 Mean : 52.61
## 3rd Qu.:160.0 3rd Qu.:660.0 3rd Qu.:108.50 3rd Qu.: 60.00
## Max. :285.0 Max. :840.0 Max. :145.00 Max. :110.00
## NA's :6 NA's :10 NA's :39 NA's :7
## PH_C PH_STUMP_L CALC_AGE CALC_ARRIVAL
## Min. : 41.0 Min. :10.00 Min. :1238 Min. : 46
## 1st Qu.:480.0 1st Qu.:25.00 1st Qu.:1901 1st Qu.:1251
## Median :540.0 Median :30.00 Median :2164 Median :1528
## Mean :542.2 Mean :30.28 Mean :2263 Mean :1566
## 3rd Qu.:620.0 3rd Qu.:36.25 3rd Qu.:2494 3rd Qu.:1851
## Max. :800.0 Max. :80.00 Max. :4590 Max. :2636
## NA's :11 NA's :39 NA's :163 NA's :55
## DH_YEAR DH_MONTH DH_QUARTER W_DIFF L_DIFF
## 2015 :176 5 :53 1: 58 Min. : 199 Min. : 20.00
## 2014 :118 9 :53 2:121 1st Qu.:1002 1st Qu.: 65.00
## 2013 : 32 6 :52 3: 85 Median :1195 Median : 80.00
## 2012 : 11 12 :37 4: 75 Mean :1238 Mean : 86.39
## 2016 : 2 3 :29 3rd Qu.:1435 3rd Qu.:100.00
## 2010 : 0 7 :25 Max. :3113 Max. :225.00
## (Other): 0 (Other):90 NA's :6 NA's :7
## C_DIFF B_YEAR B_MONTH B_QUARTER DIM_DEHORN
## Min. :-110.00 2008 : 65 3 : 31 1 : 61 Min. : 1.0
## 1st Qu.: 40.00 2010 : 47 6 : 25 2 : 61 1st Qu.: 218.8
## Median : 80.00 2009 : 20 5 : 19 3 : 34 Median : 409.5
## Mean : 78.34 2007 : 14 1 : 17 4 : 20 Mean : 494.3
## 3rd Qu.: 120.00 2006 : 8 4 : 17 NA's:163 3rd Qu.: 640.0
## Max. : 409.00 (Other): 22 (Other): 67 Max. :2070.0
## NA's :11 NA's :163 NA's :163 NA's :213
## INT_DEHORN CALC_PARITY C_YEAR C_MONTH C_QUARTER
## Min. : 0.0 0 :211 2014 : 45 2 : 25 1 : 65
## 1st Qu.: 455.0 1 : 50 2013 : 36 3 : 24 2 : 27
## Median : 601.0 2 : 58 2012 : 17 4 : 20 3 : 20
## Mean : 581.4 3 : 15 2015 : 17 1 : 16 4 : 14
## 3rd Qu.: 709.0 4 : 3 2011 : 7 12 : 14 NA's:213
## Max. :1267.0 NA's: 2 (Other): 4 (Other): 27
## NA's :2 NA's :213 NA's :213
## N_DEHORNINGS GR_DAYS G_TOTAL G_AH_W
## 1 : 0 Min. : 0.0 Min. : 0.00899 Min. : 0.00899
## 2 : 0 1st Qu.: 455.0 1st Qu.: 2.61876 1st Qu.: 2.06535
## 3 :284 Median : 601.0 Median : 3.31424 Median : 2.53264
## 4 : 48 Mean : 581.4 Mean : 5.44256 Mean : 4.09596
## 5 : 5 3rd Qu.: 709.0 3rd Qu.: 4.25376 3rd Qu.: 3.20936
## NA's: 2 Max. :1267.0 Max. :298.33333 Max. :220.33333
## NA's :2 NA's :3 NA's :3
## G_AH_L G_AH_C G_PH_W
## Min. : 0.08682 Min. : 0.4734 Min. : 0.1187
## 1st Qu.: 0.19437 1st Qu.: 0.8664 1st Qu.: 0.3550
## Median : 0.23512 Median : 1.0188 Median : 0.5006
## Mean : 0.32415 Mean : 1.5434 Mean : 1.0630
## 3rd Qu.: 0.27953 3rd Qu.: 1.2744 3rd Qu.: 0.8685
## Max. :18.33333 Max. :106.6667 Max. :48.8333
## NA's :9 NA's :13 NA's :9
## G_PH_L G_PH_C
## Min. : 0.03695 Min. : 0.05632
## 1st Qu.: 0.06957 1st Qu.: 0.73343
## Median : 0.08655 Median : 0.88962
## Mean : 0.13296 Mean : 1.32704
## 3rd Qu.: 0.10714 3rd Qu.: 1.13846
## Max. :10.00000 Max. :86.66667
## NA's :10 NA's :14
describe(df.cv.final[!df.cv.final$N_DEHORNINGS %in% c("1", "2"),])## vars n mean sd median trimmed mad min
## GENDER* 1 339 1.41 0.49 1.00 1.38 0.00 1.00
## AH_W_G 2 339 1604.89 584.27 1520.00 1560.58 447.75 5.00
## PH_W_G 3 333 391.35 262.79 315.00 350.18 173.46 47.00
## SHAVINGS_W_G 4 300 139.82 45.63 133.00 136.90 40.77 22.00
## TOTAL 5 339 2113.06 814.21 1982.00 2045.52 652.34 5.00
## AH_L 6 333 138.89 36.25 135.00 136.25 37.06 60.00
## AH_C 7 329 619.93 84.87 630.00 625.54 59.30 340.00
## AH_STUMP_L 8 300 93.72 16.56 90.00 93.16 14.83 55.00
## PH_L 9 332 52.61 16.48 50.00 51.90 14.83 10.00
## PH_C 10 328 542.17 108.45 540.00 547.01 88.96 41.00
## PH_STUMP_L 11 300 30.28 10.51 30.00 29.98 7.41 10.00
## CALC_AGE 12 176 2262.73 528.37 2163.50 2204.83 440.33 1238.00
## CALC_ARRIVAL 13 284 1566.02 437.24 1527.50 1559.60 464.80 46.00
## DH_YEAR* 14 339 5.37 0.79 6.00 5.49 0.00 3.00
## DH_MONTH* 15 339 6.88 3.13 6.00 6.86 4.45 1.00
## DH_QUARTER* 16 339 2.52 1.02 2.00 2.53 1.48 1.00
## W_DIFF 17 333 1238.23 376.67 1195.00 1212.30 326.17 199.00
## L_DIFF 18 332 86.39 29.21 80.00 84.26 29.65 20.00
## C_DIFF 19 328 78.34 58.71 80.00 78.76 59.30 -110.00
## B_YEAR* 20 176 9.31 1.72 9.00 9.51 1.48 3.00
## B_MONTH* 21 176 5.44 3.15 5.00 5.23 2.97 1.00
## B_QUARTER* 22 176 2.07 1.00 2.00 1.97 1.48 1.00
## DIM_DEHORN 23 126 494.27 363.49 409.50 443.69 297.26 1.00
## INT_DEHORN 24 337 581.37 188.27 601.00 591.79 169.02 0.00
## CALC_PARITY* 25 337 1.66 0.97 1.00 1.50 0.00 1.00
## C_YEAR* 26 126 7.27 1.26 7.00 7.38 1.48 3.00
## C_MONTH* 27 126 4.90 3.71 3.00 4.52 1.48 1.00
## C_QUARTER* 28 126 1.87 1.05 1.00 1.72 0.00 1.00
## N_DEHORNINGS* 29 337 3.17 0.42 3.00 3.07 0.00 3.00
## GR_DAYS 30 337 581.37 188.27 601.00 591.79 169.02 0.00
## G_TOTAL 31 336 5.44 17.04 3.31 3.52 1.17 0.01
## G_AH_W 32 336 4.10 12.56 2.53 2.68 0.82 0.01
## G_AH_L 33 330 0.32 1.02 0.24 0.24 0.06 0.09
## G_AH_C 34 326 1.54 5.94 1.02 1.07 0.26 0.47
## G_PH_W 35 330 1.06 3.13 0.50 0.61 0.30 0.12
## G_PH_L 36 329 0.13 0.55 0.09 0.09 0.03 0.04
## G_PH_C 37 325 1.33 4.84 0.89 0.94 0.27 0.06
## max range skew kurtosis se
## GENDER* 2.00 1.00 0.38 -1.86 0.03
## AH_W_G 4654.00 4649.00 1.20 4.29 31.73
## PH_W_G 1979.00 1932.00 2.43 9.08 14.40
## SHAVINGS_W_G 322.00 300.00 0.65 0.71 2.63
## TOTAL 6633.00 6628.00 1.36 4.94 44.22
## AH_L 285.00 225.00 0.87 1.25 1.99
## AH_C 840.00 500.00 -0.61 0.85 4.68
## AH_STUMP_L 145.00 90.00 0.33 -0.42 0.96
## PH_L 110.00 100.00 0.38 0.11 0.90
## PH_C 800.00 759.00 -0.58 1.12 5.99
## PH_STUMP_L 80.00 70.00 0.51 1.33 0.61
## CALC_AGE 4590.00 3352.00 1.27 2.49 39.83
## CALC_ARRIVAL 2636.00 2590.00 0.04 -0.11 25.95
## DH_YEAR* 7.00 4.00 -1.08 0.75 0.04
## DH_MONTH* 12.00 11.00 0.09 -1.04 0.17
## DH_QUARTER* 4.00 3.00 0.08 -1.13 0.06
## W_DIFF 3113.00 2914.00 1.02 2.72 20.64
## L_DIFF 225.00 205.00 0.89 1.52 1.60
## C_DIFF 409.00 519.00 0.49 4.24 3.24
## B_YEAR* 12.00 9.00 -0.94 1.03 0.13
## B_MONTH* 12.00 11.00 0.53 -0.64 0.24
## B_QUARTER* 4.00 3.00 0.54 -0.81 0.08
## DIM_DEHORN 2070.00 2069.00 1.58 3.18 32.38
## INT_DEHORN 1267.00 1267.00 -0.50 0.72 10.26
## CALC_PARITY* 5.00 4.00 1.24 0.48 0.05
## C_YEAR* 9.00 6.00 -0.94 1.04 0.11
## C_MONTH* 12.00 11.00 0.87 -0.75 0.33
## C_QUARTER* 4.00 3.00 0.84 -0.68 0.09
## N_DEHORNINGS* 5.00 2.00 2.33 4.82 0.02
## GR_DAYS 1267.00 1267.00 -0.50 0.72 10.26
## G_TOTAL 298.33 298.32 15.36 257.56 0.93
## G_AH_W 220.33 220.32 15.41 259.03 0.69
## G_AH_L 18.33 18.25 16.91 295.73 0.06
## G_AH_C 106.67 106.19 17.04 297.70 0.33
## G_PH_W 48.83 48.71 11.82 166.02 0.17
## G_PH_L 10.00 9.96 17.26 304.25 0.03
## G_PH_C 86.67 86.61 16.90 293.88 0.27
Next, subsets are constructed for the separate hypothesis.
We now have a subset of animals which hold the following information
summary(na.omit(df.cv[, c("CALC_ARRIVAL", "DH_QUARTER", "DIM_DEHORN", "CALC_PARITY", "C_QUARTER", "N_DEHORNINGS", "G_TOTAL", "DAYSDUE")]))## CALC_ARRIVAL DH_QUARTER DIM_DEHORN CALC_PARITY C_QUARTER
## Min. : 151 1:18 Min. : 43.0 0: 0 1:69
## 1st Qu.: 961 2:44 1st Qu.: 404.0 1:101 2:39
## Median :1319 3:54 Median : 581.0 2: 48 3:35
## Mean :1311 4:45 Mean : 598.6 3: 11 4:18
## 3rd Qu.:1683 3rd Qu.: 770.0 4: 1
## Max. :2454 Max. :1675.0
## N_DEHORNINGS G_TOTAL DAYSDUE
## 1: 2 Min. : 0.0603 Min. : 3.0
## 2:135 1st Qu.: 2.0981 1st Qu.:161.0
## 3: 20 Median : 2.4707 Median :278.0
## 4: 4 Mean : 2.6313 Mean :345.7
## 5: 0 3rd Qu.: 3.0751 3rd Qu.:495.0
## Max. :12.5602 Max. :996.0
#Histogram to explore days due
hist(df.cv[, c("DAYSDUE")],
main = "Days due",
xlab ="")
After data mining the aforementioned dataset, we decided to create a subset of COWS (number of dehornings larger than 1 with non-missing data for the following features:
- Days since arrival at the farm at dehorning (proxy for age)
- Quarter of dehorning (proxy for season)
- Days since calving at dehorning
- Calculated number of calvings
- Quarter at calving
- Number of dehornings
- Total growth
df.cv.cows <- na.omit(df.cv[!df.cv$N_DEHORNINGS == "1", c("CALC_ARRIVAL", "DH_QUARTER", "DIM_DEHORN", "CALC_PARITY", "C_QUARTER", "N_DEHORNINGS", "G_TOTAL")])
summary(df.cv.cows)## CALC_ARRIVAL DH_QUARTER DIM_DEHORN CALC_PARITY C_QUARTER
## Min. : 252 1: 56 Min. : 1.0 0: 0 1:169
## 1st Qu.:1004 2:129 1st Qu.: 267.0 1:180 2: 83
## Median :1409 3: 94 Median : 462.0 2:139 3: 71
## Mean :1398 4: 88 Mean : 531.4 3: 43 4: 44
## 3rd Qu.:1761 3rd Qu.: 717.0 4: 5
## Max. :2636 Max. :2070.0
## N_DEHORNINGS G_TOTAL
## 1: 0 Min. : 0.00899
## 2:243 1st Qu.: 2.16535
## 3:111 Median : 2.62069
## 4: 11 Mean : 3.78089
## 5: 2 3rd Qu.: 3.32012
## Max. :60.06250
#Set all number of dehornings larger then 3 equal to 3
df.cv.cows[df.cv.cows$N_DEHORNINGS == "4",]$N_DEHORNINGS <- "3"
df.cv.cows[df.cv.cows$N_DEHORNINGS == "5",]$N_DEHORNINGS <- "3"Some histograms to increase interpretation of the current dataset
op = par(mfrow=c(1, 2))
hist(df.cv.cows$G_TOTAL,
main = "Total growth (g/day)",
xlab ="")
hist(log10(df.cv.cows$G_TOTAL),
main = "Log10(Total growth (g/day))",
xlab = "")
op = par(mfrow=c(1, 1))
hist(df.cv.cows$CALC_ARRIVAL,
main = "Days since arrival at the farm",
xlab ="")
op = par(mfrow=c(2, 2))
hist(df.cv.cows[df.cv.cows$N_DEHORNINGS == "2",]$CALC_ARRIVAL,
main = "Days since arrival at the farm",
xlab ="Dehorning 2")
hist(df.cv.cows[df.cv.cows$N_DEHORNINGS == "3",]$CALC_ARRIVAL,
main = "Days since arrival at the farm",
xlab ="Dehorning 3")
hist(log10(df.cv.cows[df.cv.cows$N_DEHORNINGS == "2",]$G_TOTAL),
main = "Log10(Total growth (g/day))",
xlab ="Dehorning 2")
hist(log10(df.cv.cows[df.cv.cows$N_DEHORNINGS == "3",]$G_TOTAL),
main = "Log10(Total growth (g/day))",
xlab ="Dehorning 3")
A generalised linear model is constructed, using family="log" function.
First the baseline model is build, and next univariate models are constructed for each of the variables.
##
## Call:
## glm(formula = G_TOTAL ~ 1, family = gaussian(link = "log"), data = df.cv.cows)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -3.772 -1.616 -1.160 -0.461 56.282
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.32996 0.07722 17.22 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 31.28438)
##
## Null deviance: 11450 on 366 degrees of freedom
## Residual deviance: 11450 on 366 degrees of freedom
## AIC: 2308.1
##
## Number of Fisher Scoring iterations: 7
##
## Call:
## glm(formula = G_TOTAL ~ N_DEHORNINGS, family = gaussian(link = "log"),
## data = df.cv.cows)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -4.776 -1.866 -0.975 -0.187 55.277
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.1843 0.1090 10.863 <2e-16 ***
## N_DEHORNINGS3 0.3812 0.1508 2.527 0.0119 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 30.85296)
##
## Null deviance: 11450 on 366 degrees of freedom
## Residual deviance: 11261 on 365 degrees of freedom
## AIC: 2304
##
## Number of Fisher Scoring iterations: 7
##
## Call:
## glm(formula = G_TOTAL ~ CALC_ARRIVAL, family = gaussian(link = "log"),
## data = df.cv.cows)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -4.879 -1.745 -0.994 -0.186 55.333
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.8806835 0.1993211 9.435 <2e-16 ***
## CALC_ARRIVAL -0.0004095 0.0001545 -2.650 0.0084 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 30.77076)
##
## Null deviance: 11450 on 366 degrees of freedom
## Residual deviance: 11231 on 365 degrees of freedom
## AIC: 2303
##
## Number of Fisher Scoring iterations: 7
##
## Call:
## glm(formula = G_TOTAL ~ CALC_PARITY, family = gaussian(link = "log"),
## data = df.cv.cows)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -3.916 -1.629 -1.157 -0.446 56.180
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.3565 0.1076 12.602 <2e-16 ***
## CALC_PARITY2 0.0108 0.1621 0.067 0.947
## CALC_PARITY3 -0.2431 0.3008 -0.808 0.419
## CALC_PARITY4 -0.4608 1.0296 -0.448 0.655
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 31.44149)
##
## Null deviance: 11450 on 366 degrees of freedom
## Residual deviance: 11413 on 363 degrees of freedom
## AIC: 2312.9
##
## Number of Fisher Scoring iterations: 7
##
## Call:
## glm(formula = G_TOTAL ~ DIM_DEHORN, family = gaussian(link = "log"),
## data = df.cv.cows)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -4.336 -1.680 -1.134 -0.296 55.952
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.5197426 0.1387852 10.95 <2e-16 ***
## DIM_DEHORN -0.0003752 0.0002552 -1.47 0.142
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 31.20519)
##
## Null deviance: 11450 on 366 degrees of freedom
## Residual deviance: 11389 on 365 degrees of freedom
## AIC: 2308.2
##
## Number of Fisher Scoring iterations: 10
##
## Call:
## glm(formula = G_TOTAL ~ DH_QUARTER, family = gaussian(link = "log"),
## data = df.cv.cows)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -5.664 -1.543 -0.848 -0.093 54.338
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.7448 0.1295 13.472 < 2e-16 ***
## DH_QUARTER2 -0.5176 0.1931 -2.681 0.00768 **
## DH_QUARTER3 -0.6247 0.2272 -2.750 0.00626 **
## DH_QUARTER4 -0.3967 0.2009 -1.975 0.04908 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 30.78024)
##
## Null deviance: 11450 on 366 degrees of freedom
## Residual deviance: 11172 on 363 degrees of freedom
## AIC: 2305.1
##
## Number of Fisher Scoring iterations: 7
##
## Call:
## glm(formula = G_TOTAL ~ C_QUARTER, family = gaussian(link = "log"),
## data = df.cv.cows)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -4.153 -1.614 -1.148 -0.369 55.615
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.29250 0.11837 10.919 <2e-16 ***
## C_QUARTER2 0.19988 0.18204 1.098 0.273
## C_QUARTER3 -0.02706 0.22186 -0.122 0.903
## C_QUARTER4 -0.05751 0.27275 -0.211 0.833
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 31.40776)
##
## Null deviance: 11450 on 366 degrees of freedom
## Residual deviance: 11401 on 363 degrees of freedom
## AIC: 2312.5
##
## Number of Fisher Scoring iterations: 7
## Comparison of basemodel and single variate models
anova(baseline, glm2, glm3, glm4, glm5, glm6, glm7, test = "Chisq")## Analysis of Deviance Table
##
## Model 1: G_TOTAL ~ 1
## Model 2: G_TOTAL ~ N_DEHORNINGS
## Model 3: G_TOTAL ~ CALC_ARRIVAL
## Model 4: G_TOTAL ~ CALC_PARITY
## Model 5: G_TOTAL ~ DIM_DEHORN
## Model 6: G_TOTAL ~ DH_QUARTER
## Model 7: G_TOTAL ~ C_QUARTER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 366 11450
## 2 365 11261 1 188.843 0.01426 *
## 3 365 11231 0 29.894
## 4 363 11413 2 -181.823
## 5 365 11390 -2 23.647
## 6 363 11172 2 217.189 0.03162 *
## 7 363 11401 0 -228.457
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
We select the variables that significantly ameliorate the model fit.
summary(glm8 <- glm(G_TOTAL~ N_DEHORNINGS + CALC_ARRIVAL + DH_QUARTER,
df.cv.cows,
family = gaussian(link="log")))##
## Call:
## glm(formula = G_TOTAL ~ N_DEHORNINGS + CALC_ARRIVAL + DH_QUARTER,
## family = gaussian(link = "log"), data = df.cv.cows)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -11.161 -1.612 -0.325 0.799 50.129
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.4205722 0.1940466 12.474 < 2e-16 ***
## N_DEHORNINGS3 0.8278346 0.1512858 5.472 8.34e-08 ***
## CALC_ARRIVAL -0.0009119 0.0001585 -5.755 1.85e-08 ***
## DH_QUARTER2 -0.4567904 0.1810561 -2.523 0.0121 *
## DH_QUARTER3 -0.3355781 0.2136329 -1.571 0.1171
## DH_QUARTER4 -0.1197857 0.1750960 -0.684 0.4943
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 28.52446)
##
## Null deviance: 11450 on 366 degrees of freedom
## Residual deviance: 10297 on 361 degrees of freedom
## AIC: 2279.2
##
## Number of Fisher Scoring iterations: 9
anova(baseline, glm8, test = "Chisq")## Analysis of Deviance Table
##
## Model 1: G_TOTAL ~ 1
## Model 2: G_TOTAL ~ N_DEHORNINGS + CALC_ARRIVAL + DH_QUARTER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 366 11450
## 2 361 10297 5 1152.9 1.229e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
After exploration a one way interaction was found to significantly increase the model fit.
summary(glm9 <- glm(G_TOTAL~ N_DEHORNINGS:CALC_ARRIVAL + DH_QUARTER,
df.cv.cows,
family = gaussian(link="log")))##
## Call:
## glm(formula = G_TOTAL ~ N_DEHORNINGS:CALC_ARRIVAL + DH_QUARTER,
## family = gaussian(link = "log"), data = df.cv.cows)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -7.960 -1.702 -0.506 0.600 51.218
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.5520931 0.2030575 12.568 < 2e-16 ***
## DH_QUARTER2 -0.4855800 0.1856152 -2.616 0.00927 **
## DH_QUARTER3 -0.4630011 0.2214015 -2.091 0.03721 *
## DH_QUARTER4 -0.1738372 0.1876728 -0.926 0.35492
## N_DEHORNINGS2:CALC_ARRIVAL -0.0009053 0.0002141 -4.228 2.99e-05 ***
## N_DEHORNINGS3:CALC_ARRIVAL -0.0004665 0.0001454 -3.208 0.00146 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 29.45858)
##
## Null deviance: 11450 on 366 degrees of freedom
## Residual deviance: 10634 on 361 degrees of freedom
## AIC: 2291
##
## Number of Fisher Scoring iterations: 10
anova(baseline, glm9, test = "Chisq")## Analysis of Deviance Table
##
## Model 1: G_TOTAL ~ 1
## Model 2: G_TOTAL ~ N_DEHORNINGS:CALC_ARRIVAL + DH_QUARTER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 366 11450
## 2 361 10634 5 815.9 4.173e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
To understand the effect of the interaction, an effects plot was created
plot(effect("N_DEHORNINGS:CALC_ARRIVAL",
glm9,
#xlevels=list(CALC_ARRIVAL=900:2000),
multiline=TRUE),
main = "Interaction between number of dehornings and days at farm",
xlab = "Days since arrival",
ylab = "Total growth (g/day)")
The least square means are constructed for the final variables.
This is the overal P-value per variables
## Analysis of Deviance Table
##
## Model: gaussian, link: log
##
## Response: G_TOTAL
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 366 11450
## DH_QUARTER 3 277.75 363 11172 0.0241047 *
## N_DEHORNINGS:CALC_ARRIVAL 2 538.15 361 10634 0.0001079 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Least square means, Tukey adjusted
## NOTE: Results may be misleading due to involvement in interactions
## $lsmeans
## N_DEHORNINGS response SE df asymp.LCL asymp.UCL
## 2 2.735232 0.4143694 NA 2.032558 3.680828
## 3 5.050468 0.5283637 NA 4.114164 6.199856
##
## Results are averaged over the levels of: DH_QUARTER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast response.ratio SE df z.ratio p.value
## 2 - 3 0.5415799 0.1043803 NA -3.182 0.0015
##
## Results are averaged over the levels of: DH_QUARTER
## Tests are performed on the log scale
## NOTE: Results may be misleading due to involvement in interactions
## CALC_ARRIVAL response SE df asymp.LCL asymp.UCL
## 1397.586 3.716746 0.3253149 NA 3.130833 4.412308
##
## Results are averaged over the levels of: N_DEHORNINGS, DH_QUARTER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
## $lsmeans
## DH_QUARTER response SE df asymp.LCL asymp.UCL
## 1 4.920712 0.6835420 NA 3.747887 6.460548
## 2 3.027912 0.4495565 NA 2.263417 4.050623
## 3 3.097056 0.5771455 NA 2.149439 4.462446
## 4 4.135532 0.6008756 NA 3.110680 5.498034
##
## Results are averaged over the levels of: N_DEHORNINGS
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast response.ratio SE df z.ratio p.value
## 1 - 2 1.6251172 0.3016464 NA 2.616 0.0441
## 1 - 3 1.5888352 0.3517704 NA 2.091 0.1559
## 1 - 4 1.1898619 0.2233048 NA 0.926 0.7908
## 2 - 3 0.9776742 0.2273825 NA -0.097 0.9997
## 2 - 4 0.7321699 0.1474131 NA -1.548 0.4085
## 3 - 4 0.7488894 0.1721098 NA -1.258 0.5897
##
## Results are averaged over the levels of: N_DEHORNINGS
## P value adjustment: tukey method for comparing a family of 4 estimates
## Tests are performed on the log scale
From the previous model we learned that the moment of calving has no influence on the growth of the horn. Can we hence build a model accros sexes to increase the number of animals and get other insights.
#Make sure to drop the extemely low growth which are caused by dirty data
df.cv.all <- na.omit(df.cv[df.cv$G_TOTAL < 100,
(names(df.cv) %in% c("CALC_AGE",
"DH_QUARTER",
"GENDER",
"G_TOTAL",
"N_DEHORNINGS",
"B_QUARTER"
))])
df.cv.all[df.cv.all$N_DEHORNINGS == "4",]$N_DEHORNINGS <- "3"
df.cv.all[df.cv.all$N_DEHORNINGS == "5",]$N_DEHORNINGS <- "3"
summary(df.cv.all)## GENDER CALC_AGE DH_QUARTER B_QUARTER N_DEHORNINGS
## F:498 Min. : 58.0 1:154 1:397 1:597
## M:588 1st Qu.: 921.8 2:297 2:313 2:315
## Median :1375.0 3:275 3:183 3:174
## Mean :1552.5 4:360 4:193 4: 0
## 3rd Qu.:1964.0 5: 0
## Max. :5909.0
## G_TOTAL
## Min. : 0.01195
## 1st Qu.: 0.73166
## Median : 1.27254
## Mean : 1.99431
## 3rd Qu.: 2.74384
## Max. :33.02857
##
## Call:
## glm(formula = G_TOTAL ~ 1, family = gaussian(link = "log"), data = df.cv.all)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.9824 -1.2626 -0.7218 0.7495 31.0343
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.69030 0.03049 22.64 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 4.0144)
##
## Null deviance: 4355.6 on 1085 degrees of freedom
## Residual deviance: 4355.6 on 1085 degrees of freedom
## AIC: 4594.3
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_TOTAL ~ N_DEHORNINGS, family = gaussian(link = "log"),
## data = df.cv.all)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.4931 -0.4617 -0.1506 0.2170 29.9951
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.16885 0.07341 -2.30 0.0216 *
## N_DEHORNINGS2 1.27856 0.07862 16.26 <2e-16 ***
## N_DEHORNINGS3 1.56944 0.07868 19.95 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 2.29511)
##
## Null deviance: 4355.6 on 1085 degrees of freedom
## Residual deviance: 2485.6 on 1083 degrees of freedom
## AIC: 3989.2
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_TOTAL ~ CALC_AGE, family = gaussian(link = "log"),
## data = df.cv.all)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -4.1416 -1.0126 -0.5214 0.6007 30.9956
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.870e-01 5.778e-02 4.967 7.89e-07 ***
## CALC_AGE 2.514e-04 2.487e-05 10.107 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 3.674654)
##
## Null deviance: 4355.6 on 1085 degrees of freedom
## Residual deviance: 3983.4 on 1084 degrees of freedom
## AIC: 4499.3
##
## Number of Fisher Scoring iterations: 7
##
## Call:
## glm(formula = G_TOTAL ~ DH_QUARTER, family = gaussian(link = "log"),
## data = df.cv.all)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.943 -1.269 -0.703 0.729 30.976
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.73329 0.07763 9.446 <2e-16 ***
## DH_QUARTER2 -0.06274 0.09782 -0.641 0.521
## DH_QUARTER3 -0.01415 0.09745 -0.145 0.885
## DH_QUARTER4 -0.06832 0.09477 -0.721 0.471
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 4.022295)
##
## Null deviance: 4355.6 on 1085 degrees of freedom
## Residual deviance: 4352.1 on 1082 degrees of freedom
## AIC: 4599.5
##
## Number of Fisher Scoring iterations: 7
##
## Call:
## glm(formula = G_TOTAL ~ B_QUARTER, family = gaussian(link = "log"),
## data = df.cv.all)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.0861 -1.2225 -0.6711 0.7664 31.0625
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.59351 0.05539 10.715 < 2e-16 ***
## B_QUARTER2 0.12038 0.07827 1.538 0.12436
## B_QUARTER3 0.25739 0.08394 3.066 0.00222 **
## B_QUARTER4 0.08251 0.09176 0.899 0.36872
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 3.991866)
##
## Null deviance: 4355.6 on 1085 degrees of freedom
## Residual deviance: 4319.2 on 1082 degrees of freedom
## AIC: 4591.2
##
## Number of Fisher Scoring iterations: 7
##
## Call:
## glm(formula = G_TOTAL ~ GENDER, family = gaussian(link = "log"),
## data = df.cv.all)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.1285 -1.1594 -0.6841 0.8270 30.6865
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.4597 0.0557 8.254 4.43e-16 ***
## GENDERM 0.3913 0.0656 5.965 3.31e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 3.874997)
##
## Null deviance: 4355.6 on 1085 degrees of freedom
## Residual deviance: 4200.5 on 1084 degrees of freedom
## AIC: 4557
##
## Number of Fisher Scoring iterations: 6
## Analysis of Deviance Table
##
## Model 1: G_TOTAL ~ 1
## Model 2: G_TOTAL ~ N_DEHORNINGS
## Model 3: G_TOTAL ~ CALC_AGE
## Model 4: G_TOTAL ~ DH_QUARTER
## Model 5: G_TOTAL ~ B_QUARTER
## Model 6: G_TOTAL ~ GENDER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1085 4355.6
## 2 1083 2485.6 2 1869.99 < 2.2e-16 ***
## 3 1084 3983.4 -1 -1497.75 < 2.2e-16 ***
## 4 1082 4352.1 2 -368.77
## 5 1082 4319.2 0 32.93
## 6 1084 4200.5 -2 118.72
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_TOTAL ~ 1
## Model 2: G_TOTAL ~ N_DEHORNINGS
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1085 4355.6
## 2 1083 2485.6 2 1870 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_TOTAL ~ 1
## Model 2: G_TOTAL ~ CALC_AGE
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1085 4355.6
## 2 1084 3983.4 1 372.24 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_TOTAL ~ 1
## Model 2: G_TOTAL ~ DH_QUARTER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1085 4355.6
## 2 1082 4352.1 3 3.4655 0.8347
## Analysis of Deviance Table
##
## Model 1: G_TOTAL ~ 1
## Model 2: G_TOTAL ~ B_QUARTER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1085 4355.6
## 2 1082 4319.2 3 36.393 0.02778 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_TOTAL ~ 1
## Model 2: G_TOTAL ~ GENDER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1085 4355.6
## 2 1084 4200.5 1 155.11 2.503e-10 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Call:
## glm(formula = G_TOTAL ~ N_DEHORNINGS + CALC_AGE + GENDER, family = gaussian(link = "log"),
## data = df.cv.all)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.4434 -0.4522 -0.1100 0.2161 29.5909
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -4.727e-01 8.178e-02 -5.779 9.79e-09 ***
## N_DEHORNINGS2 1.232e+00 7.634e-02 16.140 < 2e-16 ***
## N_DEHORNINGS3 1.440e+00 7.917e-02 18.191 < 2e-16 ***
## CALC_AGE 9.878e-05 2.449e-05 4.034 5.86e-05 ***
## GENDERM 3.094e-01 4.098e-02 7.550 9.23e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 2.140349)
##
## Null deviance: 4355.6 on 1085 degrees of freedom
## Residual deviance: 2313.7 on 1081 degrees of freedom
## AIC: 3915.3
##
## Number of Fisher Scoring iterations: 6
## Analysis of Deviance Table
##
## Model: gaussian, link: log
##
## Response: G_TOTAL
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 1085 4355.6
## N_DEHORNINGS 2 1869.99 1083 2485.6 < 2.2e-16 ***
## CALC_AGE 1 37.71 1082 2447.9 2.701e-05 ***
## GENDER 1 134.20 1081 2313.7 2.408e-15 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
summary(glm8 <- glm(G_TOTAL~ N_DEHORNINGS*CALC_AGE + GENDER, df.cv.all, family = gaussian(link="log")))##
## Call:
## glm(formula = G_TOTAL ~ N_DEHORNINGS * CALC_AGE + GENDER, family = gaussian(link = "log"),
## data = df.cv.all)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.6138 -0.4478 -0.1080 0.1855 29.6003
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -3.030e-01 1.354e-01 -2.238 0.0254 *
## N_DEHORNINGS2 1.056e+00 1.467e-01 7.197 1.15e-12 ***
## N_DEHORNINGS3 1.196e+00 1.748e-01 6.842 1.30e-11 ***
## CALC_AGE -2.770e-05 9.437e-05 -0.294 0.7691
## GENDERM 3.044e-01 4.122e-02 7.386 3.01e-13 ***
## N_DEHORNINGS2:CALC_AGE 1.316e-04 9.909e-05 1.328 0.1846
## N_DEHORNINGS3:CALC_AGE 1.598e-04 1.054e-04 1.515 0.1300
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 2.138469)
##
## Null deviance: 4355.6 on 1085 degrees of freedom
## Residual deviance: 2307.4 on 1079 degrees of freedom
## AIC: 3916.4
##
## Number of Fisher Scoring iterations: 6
anova(glm8, test = "Chisq")## Analysis of Deviance Table
##
## Model: gaussian, link: log
##
## Response: G_TOTAL
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 1085 4355.6
## N_DEHORNINGS 2 1869.99 1083 2485.6 < 2.2e-16 ***
## CALC_AGE 1 37.71 1082 2447.9 2.680e-05 ***
## GENDER 1 134.20 1081 2313.7 2.342e-15 ***
## N_DEHORNINGS:CALC_AGE 2 6.30 1079 2307.4 0.2291
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
We conclude to drop the interaction in this model
## $lsmeans
## N_DEHORNINGS response SE df asymp.LCL asymp.UCL
## 1 0.848225 0.06004746 NA 0.7383341 0.9744718
## 2 2.907858 0.08378324 NA 2.7481968 3.0767956
## 3 3.581065 0.12456964 NA 3.3450495 3.8337319
##
## Results are averaged over the levels of: GENDER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast response.ratio SE df z.ratio p.value
## 1 - 2 0.2917010 0.02226705 NA -16.140 <.0001
## 1 - 3 0.2368639 0.01875336 NA -18.191 <.0001
## 2 - 3 0.8120095 0.03225660 NA -5.242 <.0001
##
## Results are averaged over the levels of: GENDER
## P value adjustment: tukey method for comparing a family of 3 estimates
## Tests are performed on the log scale
## CALC_AGE response SE df asymp.LCL asymp.UCL
## 1552.464 2.067119 0.05959925 NA 1.953546 2.187295
##
## Results are averaged over the levels of: N_DEHORNINGS, GENDER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
## $lsmeans
## GENDER response SE df asymp.LCL asymp.UCL
## F 1.770864 0.07024483 NA 1.638403 1.914035
## M 2.412936 0.07353213 NA 2.273035 2.561447
##
## Results are averaged over the levels of: N_DEHORNINGS
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast response.ratio SE df z.ratio p.value
## F - M 0.7339045 0.03007342 NA -7.55 <.0001
##
## Results are averaged over the levels of: N_DEHORNINGS
## Tests are performed on the log scale
df.cv.all.f <- na.omit(df.cv[df.cv$G_AH_W < 100,
(names(df.cv) %in% c("CALC_AGE",
"DH_QUARTER",
"GENDER",
"G_AH_W",
"N_DEHORNINGS",
"B_QUARTER"))])
summary(df.cv.all.f)## GENDER CALC_AGE DH_QUARTER B_QUARTER N_DEHORNINGS
## F:495 Min. : 58.0 1:154 1:395 1:595
## M:588 1st Qu.: 922.5 2:294 2:312 2:314
## Median :1376.0 3:275 3:183 3:140
## Mean :1553.6 4:360 4:193 4: 31
## 3rd Qu.:1965.5 5: 3
## Max. :5909.0
## G_AH_W
## Min. : 0.1148
## 1st Qu.: 0.6573
## Median : 1.1052
## Mean : 1.5912
## 3rd Qu.: 2.2017
## Max. :23.4286
df.cv.all.f[df.cv.all.f$N_DEHORNINGS == "4",]$N_DEHORNINGS <- "3"
df.cv.all.f[df.cv.all.f$N_DEHORNINGS == "5",]$N_DEHORNINGS <- "3"
#create histogram of the front horn weiht
op = par(mfrow=c(1, 2))
hist((df.cv.all.f$G_AH_W),
main = "Anterior horn weight growth (g/day)",
xlab ="")
hist(log10(df.cv.all.f$G_AH_W),
main = "Log10",
xlab ="")
##
## Call:
## glm(formula = G_AH_W ~ 1, family = gaussian(link = "log"), data = df.cv.all.f)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.4765 -0.9339 -0.4860 0.6105 21.8373
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.46451 0.02777 16.73 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 2.115181)
##
## Null deviance: 2288.6 on 1082 degrees of freedom
## Residual deviance: 2288.6 on 1082 degrees of freedom
## AIC: 3887.7
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_AH_W ~ N_DEHORNINGS, family = gaussian(link = "log"),
## data = df.cv.all.f)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.8056 -0.3797 -0.1146 0.1879 21.0315
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.30239 0.06043 -5.004 6.55e-07 ***
## N_DEHORNINGS2 1.17664 0.06565 17.923 < 2e-16 ***
## N_DEHORNINGS3 1.41791 0.06622 21.413 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 1.186824)
##
## Null deviance: 2288.6 on 1082 degrees of freedom
## Residual deviance: 1281.8 on 1080 degrees of freedom
## AIC: 3263.9
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_AH_W ~ CALC_AGE, family = gaussian(link = "log"),
## data = df.cv.all.f)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -3.0058 -0.7628 -0.3237 0.5048 21.8100
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.075e-01 5.349e-02 2.010 0.0447 *
## CALC_AGE 2.225e-04 2.383e-05 9.338 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 1.958526)
##
## Null deviance: 2288.6 on 1082 degrees of freedom
## Residual deviance: 2117.2 on 1081 degrees of freedom
## AIC: 3805.4
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_AH_W ~ DH_QUARTER, family = gaussian(link = "log"),
## data = df.cv.all.f)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.4591 -0.9438 -0.4686 0.6210 21.8098
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.51315 0.07022 7.307 5.29e-13 ***
## DH_QUARTER2 -0.05966 0.08855 -0.674 0.501
## DH_QUARTER3 -0.03149 0.08873 -0.355 0.723
## DH_QUARTER4 -0.07457 0.08591 -0.868 0.386
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 2.119406)
##
## Null deviance: 2288.6 on 1082 degrees of freedom
## Residual deviance: 2286.8 on 1079 degrees of freedom
## AIC: 3892.9
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_AH_W ~ B_QUARTER, family = gaussian(link = "log"),
## data = df.cv.all.f)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.6240 -0.9250 -0.4547 0.5987 21.8543
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.39506 0.04922 8.027 2.6e-15 ***
## B_QUARTER2 0.08046 0.07095 1.134 0.2570
## B_QUARTER3 0.19805 0.07708 2.569 0.0103 *
## B_QUARTER4 0.05870 0.08265 0.710 0.4777
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 2.108656)
##
## Null deviance: 2288.6 on 1082 degrees of freedom
## Residual deviance: 2275.2 on 1079 degrees of freedom
## AIC: 3887.4
##
## Number of Fisher Scoring iterations: 7
##
## Call:
## glm(formula = G_AH_W ~ GENDER, family = gaussian(link = "log"),
## data = df.cv.all.f)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.6154 -0.8705 -0.4967 0.6449 21.5996
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.26916 0.04917 5.474 5.45e-08 ***
## GENDERM 0.33458 0.05882 5.688 1.65e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 2.049889)
##
## Null deviance: 2288.6 on 1082 degrees of freedom
## Residual deviance: 2215.9 on 1081 degrees of freedom
## AIC: 3854.8
##
## Number of Fisher Scoring iterations: 6
## Analysis of Deviance Table
##
## Model 1: G_AH_W ~ 1
## Model 2: G_AH_W ~ N_DEHORNINGS
## Model 3: G_AH_W ~ CALC_AGE
## Model 4: G_AH_W ~ DH_QUARTER
## Model 5: G_AH_W ~ B_QUARTER
## Model 6: G_AH_W ~ GENDER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1082 2288.6
## 2 1080 1281.8 2 1006.85 < 2.2e-16 ***
## 3 1081 2117.2 -1 -835.39 < 2.2e-16 ***
## 4 1079 2286.8 2 -169.59
## 5 1079 2275.2 0 11.52
## 6 1081 2215.9 -2 59.31
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_AH_W ~ 1
## Model 2: G_AH_W ~ N_DEHORNINGS
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1082 2288.6
## 2 1080 1281.8 2 1006.9 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_AH_W ~ 1
## Model 2: G_AH_W ~ CALC_AGE
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1082 2288.6
## 2 1081 2117.2 1 171.46 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_AH_W ~ 1
## Model 2: G_AH_W ~ DH_QUARTER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1082 2288.6
## 2 1079 2286.8 3 1.8634 0.8304
## Analysis of Deviance Table
##
## Model 1: G_AH_W ~ 1
## Model 2: G_AH_W ~ B_QUARTER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1082 2288.6
## 2 1079 2275.2 3 13.382 0.09593 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_AH_W ~ 1
## Model 2: G_AH_W ~ GENDER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1082 2288.6
## 2 1081 2215.9 1 72.694 2.6e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Call:
## glm(formula = G_AH_W ~ N_DEHORNINGS + CALC_AGE + GENDER, family = gaussian(link = "log"),
## data = df.cv.all.f)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.7836 -0.3680 -0.0907 0.1817 20.7565
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -5.269e-01 6.992e-02 -7.535 1.03e-13 ***
## N_DEHORNINGS2 1.146e+00 6.479e-02 17.692 < 2e-16 ***
## N_DEHORNINGS3 1.330e+00 6.856e-02 19.403 < 2e-16 ***
## CALC_AGE 6.165e-05 2.388e-05 2.582 0.00995 **
## GENDERM 2.598e-01 3.749e-02 6.930 7.23e-12 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 1.126531)
##
## Null deviance: 2288.6 on 1082 degrees of freedom
## Residual deviance: 1214.4 on 1078 degrees of freedom
## AIC: 3209.4
##
## Number of Fisher Scoring iterations: 6
## Analysis of Deviance Table
##
## Model: gaussian, link: log
##
## Response: G_AH_W
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 1082 2288.6
## N_DEHORNINGS 2 1006.85 1080 1281.8 < 2.2e-16 ***
## CALC_AGE 1 9.18 1079 1272.6 0.004312 **
## GENDER 1 58.19 1078 1214.4 6.606e-13 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
summary(glm8 <- glm(G_AH_W~ N_DEHORNINGS*CALC_AGE +
GENDER*CALC_AGE, df.cv.all.f, family = gaussian(link="log")))##
## Call:
## glm(formula = G_AH_W ~ N_DEHORNINGS * CALC_AGE + GENDER * CALC_AGE,
## family = gaussian(link = "log"), data = df.cv.all.f)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.0286 -0.3541 -0.0859 0.1601 20.7772
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -2.718e-01 1.278e-01 -2.127 0.0336 *
## N_DEHORNINGS2 9.261e-01 1.281e-01 7.229 9.19e-13 ***
## N_DEHORNINGS3 1.046e+00 1.601e-01 6.538 9.63e-11 ***
## CALC_AGE -1.231e-04 9.075e-05 -1.357 0.1751
## GENDERM 1.697e-01 9.764e-02 1.738 0.0825 .
## N_DEHORNINGS2:CALC_AGE 1.691e-04 9.133e-05 1.852 0.0643 .
## N_DEHORNINGS3:CALC_AGE 1.921e-04 9.877e-05 1.945 0.0520 .
## CALC_AGE:GENDERM 4.389e-05 4.740e-05 0.926 0.3547
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 1.122788)
##
## Null deviance: 2288.6 on 1082 degrees of freedom
## Residual deviance: 1207.0 on 1075 degrees of freedom
## AIC: 3208.8
##
## Number of Fisher Scoring iterations: 7
anova(glm8, test = "Chisq")## Analysis of Deviance Table
##
## Model: gaussian, link: log
##
## Response: G_AH_W
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 1082 2288.6
## N_DEHORNINGS 2 1006.85 1080 1281.8 < 2.2e-16 ***
## CALC_AGE 1 9.18 1079 1272.6 0.004248 **
## GENDER 1 58.19 1078 1214.4 6.051e-13 ***
## N_DEHORNINGS:CALC_AGE 2 6.44 1076 1208.0 0.056912 .
## CALC_AGE:GENDER 1 0.97 1075 1207.0 0.352644
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
plot(effect("N_DEHORNINGS:CALC_AGE",
glm8,
#xlevels=list(CALC_ARRIVAL=900:2000),
multiline=TRUE),
main= "Interaction between number of dehornings and age",
ylab = "Anterior horn Weight growth (g/day)",
xlab = "Age")
## NOTE: Results may be misleading due to involvement in interactions
## $lsmeans
## N_DEHORNINGS response SE df asymp.LCL asymp.UCL
## 1 0.7087916 0.04899477 NA 0.6189846 0.8116286
## 2 2.3272105 0.06244837 NA 2.2079769 2.4528829
## 3 2.7203770 0.12365851 NA 2.4884937 2.9738677
##
## Results are averaged over the levels of: GENDER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast response.ratio SE df z.ratio p.value
## 1 - 2 0.3045670 0.02245851 NA -16.123 <.0001
## 1 - 3 0.2605490 0.02143324 NA -16.350 <.0001
## 2 - 3 0.8554735 0.04469707 NA -2.988 0.0079
##
## Results are averaged over the levels of: GENDER
## P value adjustment: tukey method for comparing a family of 3 estimates
## Tests are performed on the log scale
## NOTE: Results may be misleading due to involvement in interactions
## CALC_AGE response SE df asymp.LCL asymp.UCL
## 1553.608 1.649407 0.04843013 NA 1.557165 1.747113
##
## Results are averaged over the levels of: N_DEHORNINGS, GENDER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
## N_DEHORNINGS CALC_AGE response SE df asymp.LCL asymp.UCL
## 1 1553.608 0.7087916 0.04899477 NA 0.6189846 0.8116286
## 2 1553.608 2.3272105 0.06244837 NA 2.2079769 2.4528829
## 3 1553.608 2.7203770 0.12365851 NA 2.4884937 2.9738677
##
## Results are averaged over the levels of: GENDER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
## NOTE: Results may be misleading due to involvement in interactions
## $lsmeans
## GENDER response SE df asymp.LCL asymp.UCL
## F 1.464446 0.05714883 NA 1.356613 1.580851
## M 1.857728 0.06014383 NA 1.743510 1.979428
##
## Results are averaged over the levels of: N_DEHORNINGS
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast response.ratio SE df z.ratio p.value
## F - M 0.7882999 0.03244028 NA -5.78 <.0001
##
## Results are averaged over the levels of: N_DEHORNINGS
## Tests are performed on the log scale
df.cv.all.f.l <- na.omit(df.cv[df.cv$G_AH_L < 10,
(names(df.cv) %in% c("CALC_AGE",
"DH_QUARTER",
"GENDER",
"G_AH_L",
"N_DEHORNINGS",
"B_QUARTER"
))])
df.cv.all.f.l[df.cv.all.f.l$N_DEHORNINGS == "4",]$N_DEHORNINGS <- "3"
df.cv.all.f.l[df.cv.all.f.l$N_DEHORNINGS == "5",]$N_DEHORNINGS <- "3"
summary(df.cv.all.f.l)## GENDER CALC_AGE DH_QUARTER B_QUARTER N_DEHORNINGS
## F:494 Min. : 58.0 1:152 1:394 1:590
## M:584 1st Qu.: 926.2 2:294 2:308 2:314
## Median :1381.0 3:273 3:183 3:174
## Mean :1557.9 4:359 4:193 4: 0
## 3rd Qu.:1966.8 5: 0
## Max. :5909.0
## G_AH_L
## Min. :0.01447
## 1st Qu.:0.18539
## Median :0.23081
## Mean :0.22936
## 3rd Qu.:0.26828
## Max. :2.71429
op = par(mfrow=c(1, 2))
hist(df.cv.all.f.l$G_AH_L,
main = "Anterior horn length growth (mm/day)",
xlab ="")
hist(log10(df.cv.all.f.l$G_AH_L),
main = "LOG10",
xlab ="")
##
## Call:
## glm(formula = G_AH_L ~ 1, family = gaussian(link = "log"), data = df.cv.all.f.l)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.21489 -0.04397 0.00145 0.03892 2.48493
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.47247 0.01617 -91.06 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.01482908)
##
## Null deviance: 15.971 on 1077 degrees of freedom
## Residual deviance: 15.971 on 1077 degrees of freedom
## AIC: -1477.4
##
## Number of Fisher Scoring iterations: 5
##
## Call:
## glm(formula = G_AH_L ~ N_DEHORNINGS, family = gaussian(link = "log"),
## data = df.cv.all.f.l)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.17756 -0.04458 -0.00398 0.03403 2.43691
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.65007 0.02459 -67.092 < 2e-16 ***
## N_DEHORNINGS2 0.36770 0.03391 10.845 < 2e-16 ***
## N_DEHORNINGS3 0.33793 0.04060 8.324 2.56e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.01316085)
##
## Null deviance: 15.971 on 1077 degrees of freedom
## Residual deviance: 14.148 on 1075 degrees of freedom
## AIC: -1604.1
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_AH_L ~ CALC_AGE, family = gaussian(link = "log"),
## data = df.cv.all.f.l)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.17340 -0.05523 -0.00348 0.04441 2.49038
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.257e+00 3.384e-02 -37.139 < 2e-16 ***
## CALC_AGE -1.426e-04 2.166e-05 -6.585 7.09e-11 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.01415496)
##
## Null deviance: 15.971 on 1077 degrees of freedom
## Residual deviance: 15.231 on 1076 degrees of freedom
## AIC: -1526.6
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_AH_L ~ DH_QUARTER, family = gaussian(link = "log"),
## data = df.cv.all.f.l)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.21022 -0.04639 0.00033 0.03941 2.48683
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.40023 0.04005 -34.961 <2e-16 ***
## DH_QUARTER2 -0.09277 0.05102 -1.819 0.0693 .
## DH_QUARTER3 -0.08059 0.05151 -1.564 0.1180
## DH_QUARTER4 -0.08103 0.04902 -1.653 0.0986 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.01482025)
##
## Null deviance: 15.971 on 1077 degrees of freedom
## Residual deviance: 15.917 on 1074 degrees of freedom
## AIC: -1475.1
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_AH_L ~ B_QUARTER, family = gaussian(link = "log"),
## data = df.cv.all.f.l)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.21718 -0.04344 0.00136 0.03939 2.48281
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.4625120 0.0265117 -55.165 <2e-16 ***
## B_QUARTER2 -0.0127100 0.0403131 -0.315 0.753
## B_QUARTER3 -0.0370174 0.0482954 -0.766 0.444
## B_QUARTER4 -0.0007536 0.0462590 -0.016 0.987
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.01486108)
##
## Null deviance: 15.971 on 1077 degrees of freedom
## Residual deviance: 15.961 on 1074 degrees of freedom
## AIC: -1472.1
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_AH_L ~ GENDER, family = gaussian(link = "log"),
## data = df.cv.all.f.l)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.21377 -0.04322 0.00189 0.03817 2.47871
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.50506 0.02465 -61.050 <2e-16 ***
## GENDERM 0.05934 0.03262 1.819 0.0692 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.01479734)
##
## Null deviance: 15.971 on 1077 degrees of freedom
## Residual deviance: 15.921 on 1076 degrees of freedom
## AIC: -1478.8
##
## Number of Fisher Scoring iterations: 5
## Analysis of Deviance Table
##
## Model 1: G_AH_L ~ 1
## Model 2: G_AH_L ~ N_DEHORNINGS
## Model 3: G_AH_L ~ CALC_AGE
## Model 4: G_AH_L ~ DH_QUARTER
## Model 5: G_AH_L ~ B_QUARTER
## Model 6: G_AH_L ~ GENDER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1077 15.971
## 2 1075 14.148 2 1.82272 < 2.2e-16 ***
## 3 1076 15.231 -1 -1.08274 < 2.2e-16 ***
## 4 1074 15.917 2 -0.68628
## 5 1074 15.961 0 -0.04377
## 6 1076 15.921 -2 0.03937
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_AH_L ~ 1
## Model 2: G_AH_L ~ N_DEHORNINGS
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1077 15.971
## 2 1075 14.148 2 1.8227 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_AH_L ~ 1
## Model 2: G_AH_L ~ CALC_AGE
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1077 15.971
## 2 1076 15.231 1 0.73998 4.819e-13 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_AH_L ~ 1
## Model 2: G_AH_L ~ DH_QUARTER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1077 15.971
## 2 1074 15.917 3 0.053701 0.3051
## Analysis of Deviance Table
##
## Model 1: G_AH_L ~ 1
## Model 2: G_AH_L ~ B_QUARTER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1077 15.971
## 2 1074 15.961 3 0.0099274 0.8807
## Analysis of Deviance Table
##
## Model 1: G_AH_L ~ 1
## Model 2: G_AH_L ~ GENDER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1077 15.971
## 2 1076 15.921 1 0.049302 0.06795 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Call:
## glm(formula = G_AH_L ~ N_DEHORNINGS + CALC_AGE + GENDER, family = gaussian(link = "log"),
## data = df.cv.all.f.l)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.17248 -0.04321 -0.01186 0.02387 2.43545
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.217e+00 3.726e-02 -32.66 <2e-16 ***
## N_DEHORNINGS2 5.326e-01 3.442e-02 15.47 <2e-16 ***
## N_DEHORNINGS3 7.036e-01 4.976e-02 14.14 <2e-16 ***
## CALC_AGE -3.838e-04 2.975e-05 -12.90 <2e-16 ***
## GENDERM 5.222e-02 2.692e-02 1.94 0.0527 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.01080129)
##
## Null deviance: 15.971 on 1077 degrees of freedom
## Residual deviance: 11.589 on 1073 degrees of freedom
## AIC: -1815.1
##
## Number of Fisher Scoring iterations: 7
## Analysis of Deviance Table
##
## Model: gaussian, link: log
##
## Response: G_AH_L
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 1077 15.971
## N_DEHORNINGS 2 1.82272 1075 14.148 < 2e-16 ***
## CALC_AGE 1 2.51740 1074 11.630 < 2e-16 ***
## GENDER 1 0.04102 1073 11.589 0.05131 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
summary(glm8 <- glm(G_AH_L~ N_DEHORNINGS*CALC_AGE +
GENDER, df.cv.all.f.l, family = gaussian(link="log")))##
## Call:
## glm(formula = G_AH_L ~ N_DEHORNINGS * CALC_AGE + GENDER, family = gaussian(link = "log"),
## data = df.cv.all.f.l)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.18121 -0.03726 -0.00993 0.01848 2.43027
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -9.875e-01 5.089e-02 -19.405 < 2e-16 ***
## N_DEHORNINGS2 4.118e-02 8.136e-02 0.506 0.613
## N_DEHORNINGS3 -3.404e-03 1.429e-01 -0.024 0.981
## CALC_AGE -6.250e-04 5.204e-05 -12.010 < 2e-16 ***
## GENDERM 3.807e-02 2.632e-02 1.446 0.148
## N_DEHORNINGS2:CALC_AGE 4.165e-04 6.477e-05 6.431 1.91e-10 ***
## N_DEHORNINGS3:CALC_AGE 4.708e-04 7.946e-05 5.925 4.21e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.01032666)
##
## Null deviance: 15.971 on 1077 degrees of freedom
## Residual deviance: 11.060 on 1071 degrees of freedom
## AIC: -1861.5
##
## Number of Fisher Scoring iterations: 6
anova(glm8, test = "Chisq")## Analysis of Deviance Table
##
## Model: gaussian, link: log
##
## Response: G_AH_L
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 1077 15.971
## N_DEHORNINGS 2 1.82272 1075 14.148 < 2.2e-16 ***
## CALC_AGE 1 2.51740 1074 11.630 < 2.2e-16 ***
## GENDER 1 0.04102 1073 11.589 0.04625 *
## N_DEHORNINGS:CALC_AGE 2 0.52973 1071 11.060 7.259e-12 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
plot(effect("N_DEHORNINGS:CALC_AGE",
glm8,
multiline=TRUE))
## NOTE: Results may be misleading due to involvement in interactions
## $lsmeans
## N_DEHORNINGS response SE df asymp.LCL asymp.UCL
## 1 0.1433888 0.006034328 NA 0.1320364 0.1557173
## 2 0.2859036 0.005889743 NA 0.2745898 0.2976835
## 3 0.2975482 0.014046665 NA 0.2712525 0.3263931
##
## Results are averaged over the levels of: GENDER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast response.ratio SE df z.ratio p.value
## 1 - 2 0.5015286 0.02348277 NA -14.739 <.0001
## 1 - 3 0.4819011 0.03048490 NA -11.540 <.0001
## 2 - 3 0.9608646 0.04950446 NA -0.775 0.7185
##
## Results are averaged over the levels of: GENDER
## P value adjustment: tukey method for comparing a family of 3 estimates
## Tests are performed on the log scale
## NOTE: Results may be misleading due to involvement in interactions
## CALC_AGE response SE df asymp.LCL asymp.UCL
## 1557.904 0.2301958 0.00510341 NA 0.2204075 0.2404188
##
## Results are averaged over the levels of: N_DEHORNINGS, GENDER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
## N_DEHORNINGS CALC_AGE response SE df asymp.LCL asymp.UCL
## 1 1557.904 0.1433888 0.006034328 NA 0.1320364 0.1557173
## 2 1557.904 0.2859036 0.005889743 NA 0.2745898 0.2976835
## 3 1557.904 0.2975482 0.014046665 NA 0.2712525 0.3263931
##
## Results are averaged over the levels of: GENDER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
## $lsmeans
## GENDER response SE df asymp.LCL asymp.UCL
## F 0.2258551 0.005893147 NA 0.2145952 0.2377059
## M 0.2346199 0.005975375 NA 0.2231959 0.2466286
##
## Results are averaged over the levels of: N_DEHORNINGS
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast response.ratio SE df z.ratio p.value
## F - M 0.9626428 0.02533976 NA -1.446 0.1481
##
## Results are averaged over the levels of: N_DEHORNINGS
## Tests are performed on the log scale
Hypothesis 5: Factors influencing the growth of the anterior horn circumference (mm/day) in males and females
df.cv.all.f.c <- na.omit(df.cv[df.cv$G_AH_C < 100,(names(df.cv) %in% c("CALC_AGE",
"DH_QUARTER",
"GENDER",
"G_AH_C",
"N_DEHORNINGS",
"B_QUARTER"
))])
df.cv.all.f.c[df.cv.all.f.c$N_DEHORNINGS == "4",]$N_DEHORNINGS <- "3"
df.cv.all.f.c[df.cv.all.f.c$N_DEHORNINGS == "5",]$N_DEHORNINGS <- "3"
summary(df.cv.all.f.c)## GENDER CALC_AGE DH_QUARTER B_QUARTER N_DEHORNINGS
## F:488 Min. : 94.0 1:148 1:390 1:573
## M:572 1st Qu.: 932.8 2:290 2:304 2:314
## Median :1394.0 3:266 3:179 3:173
## Mean :1567.0 4:356 4:187 4: 0
## 3rd Qu.:1967.5 5: 0
## Max. :5909.0
## G_AH_C
## Min. : 0.07313
## 1st Qu.: 0.46831
## Median : 0.67344
## Mean : 0.79161
## 3rd Qu.: 0.99375
## Max. :13.28571
op = par(mfrow=c(1, 2))
hist(df.cv.all.f.c$G_AH_C,
main = "Anterior horn circumference growth(mm/day)",
xlab ="")
hist(log10(df.cv.all.f.c$G_AH_C),
main = "Log10",
xlab ="")
##
## Call:
## glm(formula = G_AH_C ~ 1, family = gaussian(link = "log"), data = df.cv.all.f.c)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.7185 -0.3233 -0.1182 0.2021 12.4941
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.23368 0.02389 -9.78 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.3792266)
##
## Null deviance: 401.6 on 1059 degrees of freedom
## Residual deviance: 401.6 on 1059 degrees of freedom
## AIC: 1983.4
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_AH_C ~ N_DEHORNINGS, family = gaussian(link = "log"),
## data = df.cv.all.f.c)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9903 -0.2164 -0.0506 0.1059 12.1381
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.72912 0.04476 -16.29 <2e-16 ***
## N_DEHORNINGS2 0.86682 0.05147 16.84 <2e-16 ***
## N_DEHORNINGS3 0.88599 0.05595 15.83 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.2670236)
##
## Null deviance: 401.60 on 1059 degrees of freedom
## Residual deviance: 282.24 on 1057 degrees of freedom
## AIC: 1613.5
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_AH_C ~ CALC_AGE, family = gaussian(link = "log"),
## data = df.cv.all.f.c)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.8552 -0.2962 -0.0942 0.1779 12.4897
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -3.238e-01 5.020e-02 -6.451 1.69e-10 ***
## CALC_AGE 5.691e-05 2.676e-05 2.127 0.0337 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.377804)
##
## Null deviance: 401.60 on 1059 degrees of freedom
## Residual deviance: 399.72 on 1058 degrees of freedom
## AIC: 1980.4
##
## Number of Fisher Scoring iterations: 6
##
## Call:
## glm(formula = G_AH_C ~ DH_QUARTER, family = gaussian(link = "log"),
## data = df.cv.all.f.c)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.7325 -0.3252 -0.1068 0.2030 12.4649
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.21618 0.06283 -3.441 0.000603 ***
## DH_QUARTER2 0.01238 0.07690 0.161 0.872176
## DH_QUARTER3 0.01874 0.07787 0.241 0.809872
## DH_QUARTER4 -0.07897 0.07661 -1.031 0.302880
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.3791523)
##
## Null deviance: 401.60 on 1059 degrees of freedom
## Residual deviance: 400.38 on 1056 degrees of freedom
## AIC: 1986.1
##
## Number of Fisher Scoring iterations: 7
##
## Call:
## glm(formula = G_AH_C ~ B_QUARTER, family = gaussian(link = "log"),
## data = df.cv.all.f.c)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.7402 -0.3195 -0.1140 0.1962 12.5024
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.25833 0.04041 -6.394 2.43e-10 ***
## B_QUARTER2 0.02823 0.06010 0.470 0.639
## B_QUARTER3 0.08094 0.06825 1.186 0.236
## B_QUARTER4 0.01414 0.07030 0.201 0.841
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.3797963)
##
## Null deviance: 401.60 on 1059 degrees of freedom
## Residual deviance: 401.06 on 1056 degrees of freedom
## AIC: 1987.9
##
## Number of Fisher Scoring iterations: 7
##
## Call:
## glm(formula = G_AH_C ~ GENDER, family = gaussian(link = "log"),
## data = df.cv.all.f.c)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.7722 -0.3295 -0.1140 0.2066 12.4356
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.32430 0.03837 -8.452 < 2e-16 ***
## GENDERM 0.16192 0.04879 3.319 0.000935 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.3755656)
##
## Null deviance: 401.60 on 1059 degrees of freedom
## Residual deviance: 397.35 on 1058 degrees of freedom
## AIC: 1974.1
##
## Number of Fisher Scoring iterations: 6
## Analysis of Deviance Table
##
## Model 1: G_AH_C ~ 1
## Model 2: G_AH_C ~ N_DEHORNINGS
## Model 3: G_AH_C ~ CALC_AGE
## Model 4: G_AH_C ~ DH_QUARTER
## Model 5: G_AH_C ~ B_QUARTER
## Model 6: G_AH_C ~ GENDER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1059 401.60
## 2 1057 282.24 2 119.357 < 2.2e-16 ***
## 3 1058 399.72 -1 -117.475 < 2.2e-16 ***
## 4 1056 400.38 2 -0.666
## 5 1056 401.06 0 -0.680
## 6 1058 397.35 -2 3.717
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_AH_C ~ 1
## Model 2: G_AH_C ~ N_DEHORNINGS
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1059 401.60
## 2 1057 282.24 2 119.36 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_AH_C ~ 1
## Model 2: G_AH_C ~ CALC_AGE
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1059 401.60
## 2 1058 399.72 1 1.8821 0.02562 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## Analysis of Deviance Table
##
## Model 1: G_AH_C ~ 1
## Model 2: G_AH_C ~ DH_QUARTER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1059 401.60
## 2 1056 400.38 3 1.216 0.3608
## Analysis of Deviance Table
##
## Model 1: G_AH_C ~ 1
## Model 2: G_AH_C ~ B_QUARTER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1059 401.60
## 2 1056 401.06 3 0.53628 0.7027
## Analysis of Deviance Table
##
## Model 1: G_AH_C ~ 1
## Model 2: G_AH_C ~ GENDER
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 1059 401.60
## 2 1058 397.35 1 4.2532 0.0007648 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Call:
## glm(formula = G_AH_C ~ N_DEHORNINGS + CALC_AGE + GENDER, family = gaussian(link = "log"),
## data = df.cv.all.f.c)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.6879 -0.1597 -0.0535 0.0764 12.0577
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -5.799e-01 5.869e-02 -9.881 < 2e-16 ***
## N_DEHORNINGS2 9.310e-01 5.251e-02 17.730 < 2e-16 ***
## N_DEHORNINGS3 1.032e+00 6.505e-02 15.860 < 2e-16 ***
## CALC_AGE -1.744e-04 3.322e-05 -5.250 1.84e-07 ***
## GENDERM 1.475e-01 3.723e-02 3.963 7.90e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.2564954)
##
## Null deviance: 401.6 on 1059 degrees of freedom
## Residual deviance: 270.6 on 1055 degrees of freedom
## AIC: 1572.8
##
## Number of Fisher Scoring iterations: 7
## Analysis of Deviance Table
##
## Model: gaussian, link: log
##
## Response: G_AH_C
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 1059 401.60
## N_DEHORNINGS 2 119.357 1057 282.24 < 2.2e-16 ***
## CALC_AGE 1 7.477 1056 274.77 6.690e-08 ***
## GENDER 1 4.166 1055 270.60 5.578e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
summary(glm8 <- glm(G_AH_C~ N_DEHORNINGS*CALC_AGE + GENDER
, df.cv.all.f.c, family = gaussian(link="log")))##
## Call:
## glm(formula = G_AH_C ~ N_DEHORNINGS * CALC_AGE + GENDER, family = gaussian(link = "log"),
## data = df.cv.all.f.c)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.6649 -0.1317 -0.0121 0.0679 12.0569
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.0541441 0.1071084 0.506 0.613308
## N_DEHORNINGS2 0.1367875 0.1251797 1.093 0.274763
## N_DEHORNINGS3 0.1996959 0.1810393 1.103 0.270256
## CALC_AGE -0.0008225 0.0001181 -6.964 5.82e-12 ***
## GENDERM 0.1389965 0.0364645 3.812 0.000146 ***
## N_DEHORNINGS2:CALC_AGE 0.0007488 0.0001238 6.046 2.06e-09 ***
## N_DEHORNINGS3:CALC_AGE 0.0007407 0.0001346 5.504 4.65e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.2472339)
##
## Null deviance: 401.60 on 1059 degrees of freedom
## Residual deviance: 260.34 on 1053 degrees of freedom
## AIC: 1535.9
##
## Number of Fisher Scoring iterations: 7
anova(glm8, test = "Chisq")## Analysis of Deviance Table
##
## Model: gaussian, link: log
##
## Response: G_AH_C
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 1059 401.60
## N_DEHORNINGS 2 119.357 1057 282.24 < 2.2e-16 ***
## CALC_AGE 1 7.477 1056 274.77 3.809e-08 ***
## GENDER 1 4.166 1055 270.60 4.047e-05 ***
## N_DEHORNINGS:CALC_AGE 2 10.265 1053 260.34 9.636e-10 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
plot(effect("N_DEHORNINGS:CALC_AGE",
glm8,
#xlevels=list(CALC_ARRIVAL=900:2000),
multiline=TRUE),
main= "Interaction between number of dehornings and age",
ylab = "Circular growth of the front horn",
xlab = "Age")
## NOTE: Results may be misleading due to involvement in interactions
## $lsmeans
## N_DEHORNINGS response SE df asymp.LCL asymp.UCL
## 1 0.3118471 0.03024745 NA 0.2578575 0.3771408
## 2 1.1559800 0.02895225 NA 1.1006049 1.2141412
## 3 1.2155102 0.06486600 NA 1.0947981 1.3495320
##
## Results are averaged over the levels of: GENDER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast response.ratio SE df z.ratio p.value
## 1 - 2 0.2697686 0.02698305 NA -13.099 <.0001
## 1 - 3 0.2565565 0.02838728 NA -12.295 <.0001
## 2 - 3 0.9510245 0.05598836 NA -0.853 0.6700
##
## Results are averaged over the levels of: GENDER
## P value adjustment: tukey method for comparing a family of 3 estimates
## Tests are performed on the log scale
## NOTE: Results may be misleading due to involvement in interactions
## CALC_AGE response SE df asymp.LCL asymp.UCL
## 1567.043 0.7595392 0.02879569 NA 0.7051466 0.8181275
##
## Results are averaged over the levels of: N_DEHORNINGS, GENDER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
## N_DEHORNINGS CALC_AGE response SE df asymp.LCL asymp.UCL
## 1 1567.043 0.3118471 0.03024745 NA 0.2578575 0.3771408
## 2 1567.043 1.1559800 0.02895225 NA 1.1006049 1.2141412
## 3 1567.043 1.2155102 0.06486600 NA 1.0947981 1.3495320
##
## Results are averaged over the levels of: GENDER
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
## $lsmeans
## GENDER response SE df asymp.LCL asymp.UCL
## F 0.7085451 0.03074814 NA 0.6507716 0.7714475
## M 0.8142034 0.03313567 NA 0.7517813 0.8818085
##
## Results are averaged over the levels of: N_DEHORNINGS
## Confidence level used: 0.95
## Intervals are back-transformed from the log scale
##
## $contrasts
## contrast response.ratio SE df z.ratio p.value
## F - M 0.8702311 0.03173252 NA -3.812 0.0001
##
## Results are averaged over the levels of: N_DEHORNINGS
## Tests are performed on the log scale
Data needs to be scaled and centered
df.cv.mfa <- na.omit(df.cv[,c("GENDER",
"N_DEHORNINGS",
"G_TOTAL",
"G_AH_W",
"G_AH_L",
"G_AH_C",
"CALC_AGE")])
df.cv.mfa$G_TOTAL <- scale(log10(df.cv.mfa$G_TOTAL), center = TRUE, scale = TRUE)
df.cv.mfa$G_AH_W <- scale(log10(df.cv.mfa$G_AH_W), center = TRUE, scale = TRUE)
df.cv.mfa$G_AH_L <- scale(log10(df.cv.mfa$G_AH_L), center = TRUE, scale = TRUE)
df.cv.mfa$G_AH_C <- scale(log10(df.cv.mfa$G_AH_C), center = TRUE, scale = TRUE)
df.cv.mfa$CALC_AGE <- scale(df.cv.mfa$CALC_AGE, center = TRUE, scale = TRUE)
df.cv.mfa[df.cv.mfa$N_DEHORNINGS == "4",]$N_DEHORNINGS <- "3"
df.cv.mfa[df.cv.mfa$N_DEHORNINGS == "5",]$N_DEHORNINGS <- "3"
describe(df.cv.mfa)## vars n mean sd median trimmed mad min max range
## GENDER* 1 1060 1.54 0.50 2.00 1.55 0.00 1.00 2.00 1.00
## N_DEHORNINGS* 2 1060 1.62 0.75 1.00 1.53 0.00 1.00 3.00 2.00
## G_TOTAL 3 1060 0.00 1.00 -0.12 -0.02 1.18 -3.09 6.49 9.58
## G_AH_W 4 1060 0.00 1.00 -0.08 -0.01 1.18 -3.09 6.81 9.90
## G_AH_L 5 1060 0.00 1.00 0.21 0.14 0.49 -5.03 8.49 13.52
## G_AH_C 6 1060 0.00 1.00 0.02 0.02 0.91 -3.57 8.20 11.77
## CALC_AGE 7 1060 0.00 1.00 -0.21 -0.13 0.90 -1.77 5.20 6.97
## skew kurtosis se
## GENDER* -0.15 -1.98 0.02
## N_DEHORNINGS* 0.74 -0.87 0.02
## G_TOTAL 0.42 0.91 0.03
## G_AH_W 0.40 1.34 0.03
## G_AH_L -0.84 8.71 0.03
## G_AH_C 0.29 4.79 0.03
## CALC_AGE 1.33 2.42 0.03
# www.sthda.com/english/articles/31-principal-component-methods-in-r-practical-guide/115-famd-factor-analysis-of-mixed-data-in-r-essentials/
cv.FAMD <- FAMD(df.cv.mfa)
Show the eigenvalues and variance and vizualize
eig.val <- get_eigenvalue(cv.FAMD)
head(eig.val)## eigenvalue variance.percent cumulative.variance.percent
## Dim.1 4.0264811 50.331014 50.33101
## Dim.2 1.5276599 19.095748 69.42676
## Dim.3 1.0291723 12.864654 82.29142
## Dim.4 0.9260278 11.575348 93.86676
## Dim.5 0.2350822 2.938528 96.80529
fviz_screeplot(cv.FAMD)
Plot the variables and overall contribution
# Define the labels of tick marks
DIM1 <- c("Anterior horn growth",
"Total growth",
"Number of dehornings",
"Anterior horn circumference",
"Anterior horn length",
"Age",
"Sex")
fviz_famd_var(cv.FAMD, repel = TRUE)
# Contribution to the first dimension
fviz_contrib(cv.FAMD, "var",
axes = 1,
x.lab = DIM1)
# Contribution to the second dimension
fviz_contrib(cv.FAMD, "var",
axes = 2)
Contribution of the quantitative variables on dimension 1 and 2
quanti.var <- get_famd_var(cv.FAMD, "quanti.var")
quanti.var ## FAMD results for quantitative variables
## ===================================================
## Name Description
## 1 "$coord" "Coordinates"
## 2 "$cos2" "Cos2, quality of representation"
## 3 "$contrib" "Contributions"
fviz_famd_var(cv.FAMD, "quanti.var", repel = TRUE,
col.var = "black")
fviz_famd_var(cv.FAMD, "quanti.var", col.var = "contrib",
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07"),
repel = TRUE)
Contribution of the qualitative variables
fviz_famd_var(cv.FAMD, "quali.var", col.var = "contrib",
gradient.cols = c("#00AFBB", "#E7B800", "#FC4E07")
)
Plot the individual observations
flabels <- c("Number of dehornings", "Sex")
fviz_ellipses(cv.FAMD, c("N_DEHORNINGS", "GENDER"),
addEllipses = TRUE,
ellipse.type = "confidence",
geom = "point",
main = "Qualitative variables"
)