Cook's d bar plot should return the threshold used in the plot

[aravindhebbali]

ols_cooksd_barplot() does not return the threshold value used in the plot.

> library(caret)
> data("Sacramento")
> lm_fit2 <- lm(price  ~ beds + baths + log(sqft), data = Sacramento)
> k <- ols_cooksd_barplot(lm_fit2)

> k$Observation
 [1]  19  48  69  74 109 121 153 154 155 156 157 158 173 278 292 294 313 315
[19] 321 322 327 329 331 332 333 334 366 382 443 457 511 543 544 547 548 549
[37] 550 551 552 553 555 599 612 705 761 781 784 801 803 805 806 807 808 809
[55] 810 811 812 813 814

> k$`Cook's Distance`
 [1] 0.004925816 0.006469155 0.009081726 0.005898704 0.007078976 0.005087501
 [7] 0.005863605 0.007676384 0.004586336 0.004469663 0.017683564 0.072050631
[13] 0.007834145 0.006855228 0.050686585 0.005360302 0.010846417 0.007199480
[19] 0.008042144 0.004714478 0.007306099 0.008606838 0.007368080 0.082314429
[25] 0.045272571 0.246924097 0.018766002 0.008591291 0.004412350 0.004904602
[31] 0.004877418 0.004500206 0.009037852 0.009439700 0.007430731 0.009794273
[37] 0.016138561 0.015583934 0.014069024 0.023726397 0.012552851 0.015113836
[43] 0.005988976 0.028129643 0.004754092 0.007761762 0.007677705 0.020385093
[49] 0.026974805 0.005075846 0.010551736 0.006427076 0.006372290 0.006065736
[55] 0.006524888 0.007945376 0.012811643 0.046742065 0.054528741

[aravindhebbali]

ols_cooksd_barplot() returns the threshold value used to classify the observations as outliers.

> library(olsrr)
> library(caret)
> data("Sacramento")
> lm_fit2 <- lm(price  ~ beds + baths + log(sqft), data = Sacramento)
> k <- ols_cooksd_barplot(lm_fit2)

> k$outliers
# A tibble: 59 × 2
   Observation `Cook's Distance`
         <int>             <dbl>
1           19       0.004925816
2           48       0.006469155
3           69       0.009081726
4           74       0.005898704
5          109       0.007078976
6          121       0.005087501
7          153       0.005863605
8          154       0.007676384
9          155       0.004586336
10         156       0.004469663
# ... with 49 more rows

> k$threshold
[1] 0.004