version 1.4.0

DESCRIPTION

@@ -1,19 +1,19 @@
 Package: lmvar
 Type: Package
 Title: Linear Regression with Non-Constant Variances
-Version: 1.3.0
+Version: 1.4.0
 Author: Posthuma Partners <info@posthuma-partners.nl> 
 Maintainer: Marco Nijmeijer <nijmeijer@posthuma-partners.nl>
 Description: Runs a linear regression in which both the expected value and the variance can vary per observation. The expected values mu follows the standard linear model mu = X_mu * beta_mu. The standard deviation sigma follows the model log(sigma) = X_sigma * beta_sigma. The package comes with two vignettes: 'Intro' gives an introduction, 'Math' gives mathematical details.
 License: GPL-3
 LazyData: TRUE
 Imports: Matrix (>= 1.2-4), matrixcalc (>= 1.0-3), maxLik (>= 1.3-4),
-        stats (>= 3.2.5), parallel (>= 3.4.0), graphics (>= 3.4.0)
+        stats (>= 3.2.5), parallel (>= 3.3.0), graphics (>= 3.3.0)
 RoxygenNote: 6.0.1
-Suggests: testthat, knitr, rmarkdown, R.rsp, MASS
+Suggests: testthat, knitr, rmarkdown, R.rsp, MASS, plotly (>= 4.7.1)
 VignetteBuilder: knitr, R.rsp
 ByteCompile: true
 NeedsCompilation: no
-Packaged: 2017-09-07 16:38:03 UTC; Marc
+Packaged: 2018-01-04 13:15:42 UTC; Marc
 Repository: CRAN
-Date/Publication: 2017-09-07 16:57:40 UTC
+Date/Publication: 2018-01-04 13:35:36 UTC

MD5

@@ -1,89 +1,96 @@
-8529d3ad0081124a394bb527ee4f4002 *DESCRIPTION
-d7b3b055a54ec3ea20025d0d9ec537f2 *NAMESPACE
-20f3d086eea94d6f505070c1c5bdba97 *NEWS.md
+737c7276afc455e1dc45ee0571f89b17 *DESCRIPTION
+ce7a76cdaa2b45d08bc84e9f85c5feb6 *NAMESPACE
+b7e04f28dbe98a892a02db51d24d750a *NEWS.md
 799ae591e93eb076cea0d9f8150235dd *R/AIC.lmvar.R
-b0137d370c4c1a01bc5ceb87d8cc6285 *R/alias.lmvar.R
+625dea541a0b4657fa96a308a8ba0f3a *R/alias.lmvar_no_fit.R
 c5a11e9f3431dd97d7e32a2df0f1fbd5 *R/beta_sigma_names.R
 6d527a34b60bbee5b434a3eb19b542a3 *R/coef.lmvar.R
-718d994f7b143739b6ef42c943db24ec *R/cv.lm.R
-949c597b1db808d95f31a63c802f49c7 *R/cv.lmvar.R
-e2a1a06a401c6301a201079f420b545c *R/dfree.R
+c5c198f98bd106402ae117a72346cd07 *R/cv.lm.R
+0583283b0f91b74fa71e2bde980aab20 *R/cv.lmvar.R
+fea491c24e8358a662f33f78cd943b36 *R/dfree.R
 39b90ad72cc9880282de9bf614d68133 *R/examples/AIC_examples.R
-bd9d7d3681defffd188f22a59df22c77 *R/examples/alias_examples.R
+cea578f43b0272013134f25920d8124e *R/examples/alias_examples.R
 13fa82f04c348bab1ab18187053eea3b *R/examples/beta_sigma_names_examples.R
 c5ae01de2d2f10f7b4bfa919e3f388a6 *R/examples/coef_examples.R
 e0010ebeb1c03a095f66a011ba8bb36d *R/examples/cv.lm_examples.R
 d7a1f1991629cf6020d032e75506be65 *R/examples/cv.lmvar_examples.R
-9d526fd9bc9b4c72c55cc5cd2614b4fa *R/examples/dfree_examples.R
+e90ab449ca7d486b0105fad6c21ba426 *R/examples/dfree_examples.R
 dd73ddf4aff15db9a6a6407af1a59577 *R/examples/fisher_examples.R
 de72e46ee5d6f79ef95e6546a123fe54 *R/examples/fitted_examples.R
 7b44ee3e2f417ae8f7b9c360962cba82 *R/examples/fwbw.lm_examples.R
-a39b7511472835d2a3d644b080897516 *R/examples/fwbw.lmvar_examples.R
+d3855d9578a0c082419605bc766848f8 *R/examples/fwbw.lmvar_no_fit_examples.R
 3c50a595183a1e9cc4caec85108ce44d *R/examples/lmvar_examples.R
+ad51d20e3090296a1e5f343be95f3f6b *R/examples/lmvar_no_fit_examples.R
 e93509e8f18367eae4c7a21672482354 *R/examples/logLik_examples.R
-02000462392443088a7b9628b773cd8a *R/examples/nobs_examples.R
+6b09738c1f2d641ebc62d78f2d335119 *R/examples/nobs_examples.R
+982d834e7e7c3fdf1a1c0190a6f46bd9 *R/examples/plot_lm_loglik_examples.R
 8bbfa2ac0a52f0be3d68985b972e0ea3 *R/examples/predict_examples.R
 014ab1c3a4fb3f2511a5df2dbe775723 *R/examples/residuals_examples.R
 ef613a5445aeaf5860f2c5fe24ea69b8 *R/examples/summary_examples.R
 2139927b488f26929811b608c885885f *R/examples/vcov_examples.R
-65c5d7d3efb959034969ce25081d0525 *R/fisher.R
-55710a0dc6e13d24c95ebd093a0cc634 *R/fitted.lmvar.R
-76d53c81c09306d3c6f9d9ab2720861c *R/fwbw.R
-056de7c2348c29e24f9427c01f92d5a6 *R/fwbw.lm.R
-b6aba946b99d5ab4fb8509b78e8c6093 *R/fwbw.lmvar.R
+a9491ae6286b3abd51633b501b4421b0 *R/fisher.R
+ca2ecc92ec74ce5e1772529bc9ec772b *R/fitted.lmvar.R
+8758cca999e5b006d4b7f9a81c7c71ef *R/fwbw.R
+db27e9d0c8e3dbe463b523ae1c987180 *R/fwbw.lm.R
+af8dc2a225899fad889071e1582e4ba2 *R/fwbw.lmvar_no_fit.R
 d97d4f2dd4613ba9af5c79e8c683d4b7 *R/gaussian_var.R
-2ad52fdbc43ab3c4307a0c7451301116 *R/lmvar.R
+5e8c6c2298c0e6f8456fb9c4ba94f7e1 *R/lmvar.R
+9579f496fa27f431642516fa710d233e *R/lmvar_no_fit.R
 8a5f8803d15359cd28f236e6a59d3d1d *R/logLik.lmvar.R
 e5cf148c174d648f080d175de6193a60 *R/make_matrix_full_rank.R
 cd8a438e6f4d4458598f14f385ca60a1 *R/make_matrix_full_rank.default.R
 b1ae7360f0cdf66e3eb8322a0e682d51 *R/make_matrix_full_rank.qr.R
-ce35ea079b9ea28fca1c413ba974790e *R/make_matrix_full_rank.sparseqr.R
+0f757f89e1585e36f19ecdfcd2be0e55 *R/make_matrix_full_rank.sparseqr.R
 f3288867bdf9afe76559cb414d136f40 *R/matrix_column_names.R
-4f63ac17353588bd3d9497d181ace9cb *R/nobs.lmvar.R
+d09df1dc05ba530d47ecef7ad48f4807 *R/nobs.lmvar_no_fit.R
 5668d333949fd9a3e2cf8099b4f1e603 *R/order_10.R
+6058d2bc592acc833c64e36b1284ede9 *R/plot_lm_loglik.R
 576c8f6c982e185744f840ebf9594d5e *R/predict.lmvar.R
-e9b6a2bd7ebec9dfe9f6221112b8db99 *R/print.cvlmvar.R
+91b511de7be5344bb6eb5ea5e7e2ed69 *R/print.cvlmvar.R
 7cd1d6d1d15de59cc06d82fe4e2ffae0 *R/print.summary_lmvar.R
 68ba53a968caa6fc52a96117a4b2a537 *R/residuals.lmvar.R
 a57fc156653d7c9ed6179fc3068774d1 *R/round_to_sd_accuracy.R
-da4359eb750fb52cb6f5913f4d34f8f3 *R/summary.lmvar.R
-7edbf99718bd27c73eef02ba0b3962bf *R/vcov.lmvar.R
-e4c8237c288027acdfbbff7b085b7e2f *README.md
-8c53f40ce779bcb91efd666d99c94436 *build/vignette.rds
+53a0a56108ded09b0576e9dce23fa7fb *R/summary.lmvar.R
+4a911bd1219a39c7ca8408ebb5c6ca0a *R/vcov.lmvar.R
+4e1b39c7885a8b4b83ae560a184a9e5b *README.md
+a198a5c97f310d864b076bae2e2b38d8 *build/vignette.rds
 f82e1918a45923e06c52f5e55835e713 *inst/doc/Intro.R
-3f28d9c0733d19e3ce0a6396832665e2 *inst/doc/Intro.Rmd
-ae5249a0a4dc5d00c470d00ddaf27121 *inst/doc/Intro.html
-30cacce24029165a47ad57078be7c082 *inst/doc/Math.ltx
-2aaff8ee1e29c7dfd486da3ab9a9b65f *inst/doc/Math.pdf
+54f125c9c4dd72e6226fdccdaa14dee9 *inst/doc/Intro.Rmd
+32b0eb5b34cb60bd4ab4eb9e37d62cf7 *inst/doc/Intro.html
+d14af0a8f0cc7cd7096c4d68cff40a9c *inst/doc/Math.ltx
+07024a16f969facb22ae41e3e0003d60 *inst/doc/Math.pdf
 cdc8775371ef46701cbbfdd098441d22 *man/AIC.lmvar.Rd
-40be9e23383ce6cfb87f6158572eab2e *man/alias.lmvar.Rd
+9155d6d400a240a721c21b9fcd3f2600 *man/alias.lmvar_no_fit.Rd
 96870ffeb6e33b5476b9014730e99d0f *man/beta_sigma_names.Rd
 080d71de1dad4ff42e68810ec5e3c56f *man/coef.lmvar.Rd
-eda5d04e9440ca14271b251dbc23b0a4 *man/cv.lm.Rd
-194343163cc18b220ac7779ebb479cf8 *man/cv.lmvar.Rd
-d728f54df747c77aa92e6dc2c20e1897 *man/dfree.Rd
-a5e5ccd1c0fafcdaee3bbeb7cab47cd5 *man/fisher.Rd
-e7644d21028376eabaaa0d6be23001d3 *man/fitted.lmvar.Rd
-48b61963e666087c40453b34bcd2f59e *man/fwbw.Rd
-2767d53d216092f2fcc458bd6d698bf2 *man/fwbw.lm.Rd
-618a816bc8e66bdaa03bbdca3c91e485 *man/fwbw.lmvar.Rd
-ff64c128f2e97962b98c5bcac72bee22 *man/lmvar.Rd
+21f8b67c390611bbaba18e7e8a08b389 *man/cv.lm.Rd
+13cf16bd81673c937478dc3fccc53602 *man/cv.lmvar.Rd
+1f23082551603052841421c699e04695 *man/dfree.Rd
+2841cd34efa9a37ee83c2f831e790859 *man/fisher.Rd
+731b8ccbc9a0ffd0e2d1f879a7b34d1c *man/fitted.lmvar.Rd
+8ccc2d718aa8d9918fbccc7b5a20ec50 *man/fwbw.Rd
+63139890a9ffd9f1c93e1c814f8341e0 *man/fwbw.lm.Rd
+2e95d75eadf3cd8f07698f5fc0f327a9 *man/fwbw.lmvar_no_fit.Rd
+f4be60c654357a66c092e6c6db62d032 *man/lmvar.Rd
+54d6a4f3a24741f8b8188763d8c1d02e *man/lmvar_no_fit.Rd
 d3041936af6e96cdc0ae4464e0f52796 *man/logLik.lmvar.Rd
-c33fe1105392aeab24e540c8f09459a4 *man/nobs.lmvar.Rd
+092345fcff1e6e8f50a906608fe1e0c6 *man/nobs.lmvar_no_fit.Rd
+ec03ba6d5de9d5be935003325e2103b8 *man/plot_lm_loglik.Rd
 f7d8f046405077977aeeff7a3e9ab0e5 *man/predict.lmvar.Rd
 ae4ae2b043239120febca0d18f449c99 *man/print.cvlmvar.Rd
 080d695ec1395b492167dbf7e898c71a *man/print.summary_lmvar.Rd
 9b35a0b154bc37c64d22c063fe84850e *man/residuals.lmvar.Rd
-7fca9ba8b3a74e8447ce5e544f35fb7e *man/summary.lmvar.Rd
-869b39bcf7108f031d972e1883df6a22 *man/vcov.lmvar.Rd
+1c0ecff7c342cb941990a9e41ff2eeaa *man/summary.lmvar.Rd
+4fc6426dd6c460f89da7c2d655cd3010 *man/vcov.lmvar.Rd
 cbc022948d6c0009e705b9f0111937cf *tests/testthat.R
 2240bdeb95bf161a84eb0780c4f55a4f *tests/testthat/helper.R
 6a3c1e1754a425df52885f76e6efe33b *tests/testthat/test_cross_validations.R
-9181c1bb3ff9b24d1d5da2dcf1a0ce1e *tests/testthat/test_extractors.R
-f9c3bdb716810b6674fd79b18a44c2e4 *tests/testthat/test_fwbw.R
+0904e44a027ff53c93eb48cd7b7ff128 *tests/testthat/test_extractors.R
+da092f43aa741122767a3b28f9fb6475 *tests/testthat/test_fwbw.R
 662b803897ac92828a7d5368f62f7e45 *tests/testthat/test_gradient_and_hessian.R
-1fc85f6a930d7cb215c91afc563c4b8d *tests/testthat/test_lmvar.R
+d3438cc41b003a1a791f90464d9310b3 *tests/testthat/test_lmvar.R
+c1c85a254392910637cb73094a56e120 *tests/testthat/test_lmvar_no_fit.R
 2de1a3ebd02098ee93f67c401f92d281 *tests/testthat/test_predict.R
-3f28d9c0733d19e3ce0a6396832665e2 *vignettes/Intro.Rmd
-30cacce24029165a47ad57078be7c082 *vignettes/Math.ltx
+54f125c9c4dd72e6226fdccdaa14dee9 *vignettes/Intro.Rmd
+d14af0a8f0cc7cd7096c4d68cff40a9c *vignettes/Math.ltx
 3bdf41fcf4a9097ce1e06f5367c48ea9 *vignettes/bibliography.bib

NAMESPACE

@@ -1,13 +1,13 @@
 # Generated by roxygen2: do not edit by hand
 
 S3method(AIC,lmvar)
-S3method(alias,lmvar)
+S3method(alias,lmvar_no_fit)
 S3method(coef,lmvar)
 S3method(fitted,lmvar)
 S3method(fwbw,lm)
-S3method(fwbw,lmvar)
+S3method(fwbw,lmvar_no_fit)
 S3method(logLik,lmvar)
-S3method(nobs,lmvar)
+S3method(nobs,lmvar_no_fit)
 S3method(predict,lmvar)
 S3method(print,cvlmvar)
 S3method(print,summary_lmvar)
@@ -21,5 +21,6 @@ export(dfree)
 export(fisher)
 export(fwbw)
 export(lmvar)
+export(lmvar_no_fit)
 importFrom(stats,alias)
 importFrom(stats,nobs)

NEWS.md

@@ -1,3 +1,32 @@
+Version 1.4.0
+-------------
+
+* Introduce the class 'lmvar_no_fit'. This class is like the class 'lmvar', but without members that
+are the result of a model fit. The constructor of an object of class 'lmvar_no_fit' is the function 'lmvar_no_fit()'.  
+  The class 'lmvar' is an extension of the class 'lmvar_no_fit'. This means that wherever an object of class 'lmvar_no_fit' is required, an object of class 'lmvar' can be used as well.   
+  The class 'lmvar_no_fit' was motivated by situations in which 'lmvar()' does not converge for a model with many degrees of freedom
+and one wants to resort to 'fwbw()' to obtain a subset of degrees of freedom which does converge.  
+
+* Modify appropriate functions (such as 'dfree()' and 'fwbw.lmvar()') such that they take a 'lmvar_no_fit' object as input.
+
+* Add control option 'remove_df_sigma' to function 'lmvar()'. In cases where the 'lmvar' fit does not converge, switching on this option
+may restore convergence. Update both vingettes to give information about this option. 
+
+* Change the algorithm of the functions 'fwbw.lm()' and 'fwbw.lmvar_no_fit()'. The change makes the insert and remove step of the algorithm symmetric. Rather than inserting degrees of freedom one-by-one, it will attempt to insert a percentage of the left-out degrees of freedom, just like the algorithm attempts to remove a percentage of the degrees of freedom that are included.  
+  This makes it feasible to run the functions with the option `fw = TRUE`, even for models with potentially many degrees of freedom. However, the outcome of the functions can be different compared to previous versions of the 'lmvar' package.
+
+* Rename control option 'running_diagnostics' to 'monitor' for function 'lmvar()'. This is consistent with the function 'fwbw()'. 
+
+* Make calculation of $\beta_\mu$ in 'lmvar()' more robust.
+
+* Minor change in print format of object of class 'cvlmvar' to be more consistent with accepted terminology.
+
+* Fix bug which can cause intercept term to be removed when matrix of class 'Matrix' is made full-rank. 
+
+* Fix bug in 'lmvar()' which can cause wrong beta to be reported during run when control option 'monitor' is set to TRUE.
+
+* Fix bug in 'fwbw.lmvar_no_fit()' which causes an error when an initial estimate for $\beta_\sigma$ was specified in `object`.
+
 Version 1.3.0
 -------------
 

R/alias.lmvar.R → R/alias.lmvar_no_fit.R

@@ -4,7 +4,7 @@
 #' @description Returns the columns present in the user-specified model-matrices \eqn{X_\mu} and \eqn{X_\sigma} that were removed by
 #' \code{lmvar} to make the matrices full-rank.
 #'
-#' @param object Object of class 'lmvar'
+#' @param object Object of class 'lmvar_no_fit' (hence it can also be of class 'lmvar')
 #' @param mu Boolean, specifies whether the aliased columns from the model matrix \eqn{X_\mu} must be returned
 #' @param sigma Boolean, specifies whether the aliased columns from the model matrix \eqn{X_\sigma} must be returned
 #' @param ... Additional arguments, not used in the current implementation
@@ -25,7 +25,7 @@
 #'
 #' @example R/examples/alias_examples.R
 #'
-alias.lmvar <- function( object, mu = TRUE, sigma = TRUE, ...){
+alias.lmvar_no_fit <- function( object, mu = TRUE, sigma = TRUE, ...){
 
   aliased_mu = character()
   if (mu){

R/cv.lm.R

@@ -27,8 +27,8 @@
 #' }
 #' \item \code{MSE_sqrt} a list with two items
 #' \itemize{
-#' \item \code{mean} the sample mean of the square root of the mean squared prediction error over the k folds
-#' \item \code{sd} the sample standard deviation of the square root of the mean squared prediction error
+#' \item \code{mean} the sample mean of the root mean squared prediction error over the k folds
+#' \item \code{sd} the sample standard deviation of the root mean squared prediction error
 #' over the k folds
 #' }
 #' \item \code{KS_distance} a list with two items

R/cv.lmvar.R

@@ -35,8 +35,8 @@
 #' }
 #' \item \code{MSE_sqrt} a list with two items
 #' \itemize{
-#' \item \code{mean} the sample mean of the square root of the mean squared prediction error over the k folds
-#' \item \code{sd} the sample standard deviation of the square root of the mean squared prediction error
+#' \item \code{mean} the sample mean of the root mean squared prediction error over the k folds
+#' \item \code{sd} the sample standard deviation of the root mean squared prediction error
 #' over the k folds
 #' }
 #' \item \code{KS_distance} a list with two items

R/dfree.R

@@ -3,7 +3,7 @@
 #' @description Degrees of freedom for the model in an object of class 'lmvar'. The degrees of freedom are defined as the rank of the
 #' model matrix \eqn{X_\mu} for the expectation values, plus the rank of the model matrix \eqn{X_\sigma} for the standard deviations.
 #'
-#' @param object Object of class 'lmvar'
+#' @param object Object of class 'lmvar_no_fit' (hence it can also be of class 'lmvar')
 #' @param mu Boolean, specifies whether the degrees of freedom for the model for the expectation values must be included.
 #' @param sigma Boolean, specifies whether the degrees of freedom for the model for the standard deviations must be included.
 #' @param ... Additional arguments, not used in the current implementation
@@ -27,8 +27,8 @@
 
 dfree <- function( object, mu = TRUE, sigma = TRUE, ...){
 
-  if (class(object) != 'lmvar'){
-    stop("Object must be an 'lmvar' object")
+  if (!('lmvar_no_fit' %in% class(object))){
+    stop("Object must be an 'lmvar_no_fit' (or 'lmvar') object")
   }
 
   if (mu & sigma){

R/examples/alias_examples.R

@@ -29,3 +29,7 @@ alias(fit, sigma = FALSE)
 
 # Only return the aliased colums in the model matrix for the standard deviations
 alias(fit, mu = FALSE)
+
+# It also works on an object of class 'lmvar_no_fit'
+no_fit = lmvar_no_fit(y, X, X_s)
+alias(no_fit, mu = FALSE)

R/examples/dfree_examples.R

@@ -21,3 +21,7 @@ dfree(fit, sigma = FALSE)
 
 # The degrees of freedom of the standard deviations are
 dfree(fit, mu = FALSE)
+
+# Function also works on object of class 'lmvar_no_fit'
+no_fit = lmvar_no_fit(attenu$accel, X, X_s)
+dfree(no_fit)

R/examples/fwbw.lmvar_examples.R → R/examples/fwbw.lmvar_no_fit_examples.R

@@ -41,3 +41,7 @@ names(coef(fwbw$object))
 fwbw = fwbw(fit, BIC, fw = TRUE)
 names(coef(fwbw$object))
 
+# It also works on an object of class 'lmvar_no_fit'
+no_fit = lmvar_no_fit(y, X_mu, X_sigma)
+fwbw( no_fit, AIC, control = list(monitor = TRUE))
+

R/examples/lmvar_no_fit_examples.R

@@ -0,0 +1,25 @@
+# As example we sue the dataset 'iris' from the library 'datasets'
+library(datasets)
+
+# Create the model matrix for both the expected values and the standard deviations
+X = model.matrix( ~ Species - 1, data = iris)
+
+# Take as response variabe the variable Sepal.length
+y = iris$Sepal.Length
+
+# Construct a 'lmvar_no_fit' object
+no_fit = lmvar_no_fit( y, X, X)
+
+# The following functions all work on such an object
+nobs(no_fit)
+dfree(no_fit)
+alias(no_fit)
+
+# You can also supply 'lmvar' arguments
+no_fit = lmvar_no_fit( y, X[,-1], X[,-1], intercept_mu = FALSE, intercept_sigma = FALSE)
+dfree(no_fit)
+
+# Some (most) arguments have no effect except that they are stored in the 'lmvar_no_fit'
+# object
+no_fit = lmvar_no_fit( y, X, X, control = list( slvr_log = TRUE, remove_df_sigma = TRUE))
+no_fit$control

R/examples/nobs_examples.R

@@ -18,3 +18,8 @@ nobs(fit)
 
 # Check that this is equal to the number of observations in the dataset
 nobs(fit) == nrow(attenu)
+
+# Function also works on object of class 'lmvar_no_fit'
+no_fit = lmvar_no_fit(attenu$accel, X, X_s)
+nobs(no_fit)
+

R/examples/plot_lm_loglik_examples.R

@@ -0,0 +1,36 @@
+\dontrun{
+
+# Carry out a linear regression with the 'iris' data set
+fit = lm( Petal.Length ~ Species, data = iris, x = TRUE, y = TRUE)
+X = fit$x
+y = fit$y
+
+# We center the plot at the maximum-likelihood
+beta_or = coef(fit)
+
+# Plot the maximum log-likelihood
+lmvar:::plot_lm_loglik( y, X, beta_or = beta_or, beta_x = "(Intercept)",
+                        beta_y = "Speciesversicolor")
+
+# Plot against the two species
+lmvar:::plot_lm_loglik( y, X, beta_or = beta_or, beta_x = "Speciesversicolor",
+                        beta_y = "Speciesvirginica")
+
+# Increase the resolution
+lmvar:::plot_lm_loglik( y, X, beta_or = beta_or, beta_x = "Speciesversicolor",
+                        beta_y = "Speciesvirginica", plot_points = 40)
+
+# Remove the intercept term from the model matrix and fit again
+XX = X[,-1]
+fit = lm( y ~ . - 1, data = as.data.frame(XX))
+
+# Estimate the effect of adding an intercept term in a quadratic approximation and compare
+# with exact result
+beta_or = c( 0, coef(fit))
+lmvar:::plot_lm_loglik( y, X, beta_or = beta_or, beta_x = 1, beta_y = "Speciesversicolor",
+                        add_qa = TRUE, plot_points = 40, plot_width = 5)
+
+# Note that in the last case the quadratic approximation has no maximum. Hence the beta for
+# "Speciesvirginica" is kept at beta_or[3] in the calculation of the surface of the
+# quadratic approximation.
+}

R/fisher.R

@@ -29,7 +29,7 @@
 #' @seealso \code{\link{vcov.lmvar}} calculates the covariance matrix for the maximum-likelihood estimators of
 #' \eqn{\beta_\mu} and \eqn{\beta_\mu}
 #'
-#' \code{\link{nobs.lmvar}} for the number of observations in an object of class 'lmvar'
+#' \code{\link{nobs.lmvar_no_fit}} for the number of observations in an object of class 'lmvar'
 #'
 #' \code{\link{coef.lmvar}} for the coefficients \eqn{\beta_\mu} and \eqn{\beta_\sigma}
 #'

R/fitted.lmvar.R

@@ -42,7 +42,7 @@
 #'
 #' If \code{log = TRUE}, \code{fitted.lmvar} returns estimators and intervals for \eqn{e^Y}.
 #'
-#' Confidence intervals are calculated under teh assumption of asymptotic normality. This assumption holds when the number of observations is
+#' Confidence intervals are calculated under the assumption of asymptotic normality. This assumption holds when the number of observations is
 #' large. Intervals must be treated cautiously in case of a small number of observations.
 #' Intervals can also be unreliable if
 #' \code{object} was created with a constraint on the minimum values of the standard deviations sigma.