version 1.1.0

DESCRIPTION

@@ -1,18 +1,18 @@
 Package: lmvar
 Type: Package
 Title: Linear Regression with Non-Constant Variances
-Version: 1.0.0
+Version: 1.1.0
 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), nleqslv (>= 3.0.3),
+Imports: Matrix (>= 1.2-4), matrixcalc (>= 1.0-3), maxLik (>= 1.3-4),
         stats (>= 3.2.5)
 RoxygenNote: 6.0.1
-Suggests: testthat, knitr, rmarkdown, R.rsp
+Suggests: testthat, knitr, rmarkdown, R.rsp, MASS
 VignetteBuilder: knitr, R.rsp
 NeedsCompilation: no
-Packaged: 2017-02-16 16:39:12 UTC; Marc
+Packaged: 2017-03-29 07:29:11 UTC; Marc
 Repository: CRAN
-Date/Publication: 2017-02-17 14:38:14
+Date/Publication: 2017-03-29 12:17:35 UTC

MD5

@@ -1,66 +1,66 @@
-0da3dd6021303151430b92d13b01bc04 *DESCRIPTION
-41b7f05cd32d8769ef2208ad6368d7e9 *NAMESPACE
+2edd2292dd3f51dff8535729b90bb4dd *DESCRIPTION
+52973f7e6426bc4359492c3024b599e1 *NAMESPACE
+b65a0f5077bda01695c98aa4bd29aa30 *NEWS.md
 799ae591e93eb076cea0d9f8150235dd *R/AIC.lmvar.R
 b0137d370c4c1a01bc5ceb87d8cc6285 *R/alias.lmvar.R
 c5a11e9f3431dd97d7e32a2df0f1fbd5 *R/beta_sigma_names.R
-08618c12995a54734f49b0735c535257 *R/coef.lmvar.R
-010573949388442cd98049ecc6b0e7b2 *R/df.residual.lmvar.R
-a9d9d476093ad924986c191b5241769b *R/dfree.R
+6d527a34b60bbee5b434a3eb19b542a3 *R/coef.lmvar.R
+e2a1a06a401c6301a201079f420b545c *R/dfree.R
 39b90ad72cc9880282de9bf614d68133 *R/examples/AIC_examples.R
 bd9d7d3681defffd188f22a59df22c77 *R/examples/alias_examples.R
 13fa82f04c348bab1ab18187053eea3b *R/examples/beta_sigma_names_examples.R
 c5ae01de2d2f10f7b4bfa919e3f388a6 *R/examples/coef_examples.R
-ee365be6d30d6714e4a6454e6a0bfef7 *R/examples/df.residual_examples.R
 9d526fd9bc9b4c72c55cc5cd2614b4fa *R/examples/dfree_examples.R
 dd73ddf4aff15db9a6a6407af1a59577 *R/examples/fisher_examples.R
-698ac483d76f91ef2a49fc965577ee33 *R/examples/fitted_examples.R
-8cb1e5039896f1872911159fc875f758 *R/examples/lmvar_examples.R
+de72e46ee5d6f79ef95e6546a123fe54 *R/examples/fitted_examples.R
+f077a55e199e58dae1d27d25335db884 *R/examples/lmvar_examples.R
 e93509e8f18367eae4c7a21672482354 *R/examples/logLik_examples.R
 02000462392443088a7b9628b773cd8a *R/examples/nobs_examples.R
-f28168301e48d6a216067c9e6f46320a *R/examples/predict_examples.R
+8bbfa2ac0a52f0be3d68985b972e0ea3 *R/examples/predict_examples.R
 014ab1c3a4fb3f2511a5df2dbe775723 *R/examples/residuals_examples.R
-3b54dd7601fce2246ecfe4fbebaf345f *R/examples/summary_examples.R
+ef613a5445aeaf5860f2c5fe24ea69b8 *R/examples/summary_examples.R
 2139927b488f26929811b608c885885f *R/examples/vcov_examples.R
 65c5d7d3efb959034969ce25081d0525 *R/fisher.R
-bb4e182ff8110e449ab82117c67b5f6f *R/fitted.lmvar.R
+f7f5e1b02453dee36675c0b93c29e044 *R/fitted.lmvar.R
 d97d4f2dd4613ba9af5c79e8c683d4b7 *R/gaussian_var.R
-ae576df528a3f5b30cac7df2df9dce68 *R/lmvar.R
+e52430dee384e8fc8c25fe9e2dc1e615 *R/lmvar.R
 8a5f8803d15359cd28f236e6a59d3d1d *R/logLik.lmvar.R
-11a615ab05ce7c6a45d5905a064587b6 *R/make_matrix_full_rank.R
-c6eb72596cb95e6d1f12968a110a1fcb *R/matrix_column_names.R
+d3f6d391dc6c4b1d11d42cc305cc14c7 *R/make_matrix_full_rank.R
+f3288867bdf9afe76559cb414d136f40 *R/matrix_column_names.R
 4f63ac17353588bd3d9497d181ace9cb *R/nobs.lmvar.R
-f37f6f29c93effe8ebfdc3998dc28a20 *R/predict.lmvar.R
-af17a34e533280763874b7faa31208a1 *R/print.summary_lmvar.R
+51408820b0c6ceb5a05ba9572bcf563b *R/predict.lmvar.R
+6e05efb82387a3ee531c8c2f5383caa0 *R/print.summary_lmvar.R
 68ba53a968caa6fc52a96117a4b2a537 *R/residuals.lmvar.R
-6f3aa4fe0edbf10588592d9bef786368 *R/summary.lmvar.R
-b07e75a5335b4375cedc45d4e876594f *R/vcov.lmvar.R
-058e5cb3b2ffde63a0e073c229b7ca24 *README.md
-9a51ea3e128c9edff4d3fe1f448b53a5 *build/vignette.rds
-828d6a905ed48145c432c23f00ef645c *inst/doc/Intro.R
-475cfafe2a4d03be48ab649bab4d5f81 *inst/doc/Intro.Rmd
-7fff116f96908a44f030b786be788671 *inst/doc/Intro.html
-0d2d2347fc49b246c9f6607324c60005 *inst/doc/Math.ltx
-604b46aa4979321bd4c90566bfa4cc44 *inst/doc/Math.pdf
+84f68669bf255b36289779722a24c76a *R/summary.lmvar.R
+7edbf99718bd27c73eef02ba0b3962bf *R/vcov.lmvar.R
+4b781b692f552a9f22d261ef010b6d5a *README.md
+c579eb4b11dc36f29724cfa28a9f628b *build/vignette.rds
+f82e1918a45923e06c52f5e55835e713 *inst/doc/Intro.R
+2439946a1cce9cce99a20fe0c0a740d1 *inst/doc/Intro.Rmd
+4eac84742c5496ecfdee8fee837b8df5 *inst/doc/Intro.html
+2e88d78c1f638f78fd981a4d677b730e *inst/doc/Math.ltx
+b1fd4dff43279b5da2db5a2e54129aed *inst/doc/Math.pdf
 cdc8775371ef46701cbbfdd098441d22 *man/AIC.lmvar.Rd
 40be9e23383ce6cfb87f6158572eab2e *man/alias.lmvar.Rd
 96870ffeb6e33b5476b9014730e99d0f *man/beta_sigma_names.Rd
-dc231983bd8fc1c8234230199df56ebc *man/coef.lmvar.Rd
-ae69a6216348af4e1787e9550574a03e *man/df.residual.lmvar.Rd
-8a46914afe8717717af7868eb56fe898 *man/dfree.Rd
+080d71de1dad4ff42e68810ec5e3c56f *man/coef.lmvar.Rd
+d728f54df747c77aa92e6dc2c20e1897 *man/dfree.Rd
 a5e5ccd1c0fafcdaee3bbeb7cab47cd5 *man/fisher.Rd
-f84bf4c6fc34202b0f52fc560ac08b84 *man/fitted.lmvar.Rd
-9105d49018403123fc7fe8430ec2299a *man/lmvar.Rd
+0ae10d96bcc680bbf04b1011fe64662e *man/fitted.lmvar.Rd
+992319913416906156e541f1a92addcd *man/lmvar.Rd
 d3041936af6e96cdc0ae4464e0f52796 *man/logLik.lmvar.Rd
 c33fe1105392aeab24e540c8f09459a4 *man/nobs.lmvar.Rd
-85f6305560b23fac8565e52f25fc606a *man/predict.lmvar.Rd
+91e5550ddd2e0347bf055aca01310f43 *man/predict.lmvar.Rd
 080d695ec1395b492167dbf7e898c71a *man/print.summary_lmvar.Rd
 9b35a0b154bc37c64d22c063fe84850e *man/residuals.lmvar.Rd
-3001d4fdbb5a458419ddcac4c7e161f7 *man/summary.lmvar.Rd
-103ad2096a2c2810478be7d856ea2b32 *man/vcov.lmvar.Rd
-bde5eb3a375b2d0298b8c17d84eda405 *tests/create_test_data.R
+f40cafc12225a417527d4e20d3940766 *man/summary.lmvar.Rd
+869b39bcf7108f031d972e1883df6a22 *man/vcov.lmvar.Rd
+2240bdeb95bf161a84eb0780c4f55a4f *tests/create_test_data.R
 b5e9bbfbc1b319bc3baec90d163e1426 *tests/testthat.R
 9181c1bb3ff9b24d1d5da2dcf1a0ce1e *tests/testthat/test_extractors.R
-3dc4e0f80fc5c12cef9aadd1a98116f7 *tests/testthat/test_lmvar.R
-18b7f56763d877e27e3f5083153c384f *tests/testthat/test_predict.R
-475cfafe2a4d03be48ab649bab4d5f81 *vignettes/Intro.Rmd
-0d2d2347fc49b246c9f6607324c60005 *vignettes/Math.ltx
+e6bfe10c846af198d038690951ced931 *tests/testthat/test_gradient_and_hessian.R
+42f498d5a3ca3706de5df38c4d4047c5 *tests/testthat/test_lmvar.R
+2de1a3ebd02098ee93f67c401f92d281 *tests/testthat/test_predict.R
+2439946a1cce9cce99a20fe0c0a740d1 *vignettes/Intro.Rmd
+2e88d78c1f638f78fd981a4d677b730e *vignettes/Math.ltx
+1f1187e559041319b24a3d5459fd64c7 *vignettes/bibliography.bib

NAMESPACE

@@ -3,7 +3,6 @@
 S3method(AIC,lmvar)
 S3method(alias,lmvar)
 S3method(coef,lmvar)
-S3method(df.residual,lmvar)
 S3method(fitted,lmvar)
 S3method(logLik,lmvar)
 S3method(nobs,lmvar)

NEWS.md

@@ -0,0 +1,27 @@
+Version 1.1.0
+-------------
+
+* To find the maximum likelihood, a set of non-linear equations must be solved. This is done by the function 'maxNR' from the package 'maxLik'. It results in faster and more robust solves, compared to the previous version of 'lmvar'. 
+
+* It is possible to pass on options to 'maxNR' from the call to 'lmvar'. For this, 'lmvar' has a new argument 'slvr_options'.
+
+* It is possible to request the result log of 'maxNR'. For this, 'lmvar' has a new argument 'slvr_log'.
+
+* It is possible to suppress the intercept terms in the model for mu and the model for log sigma. For this, 'lmvar' has the new arguments 'intercept_mu' and 'intercept_sigma'.
+
+* The functions 'fitted.lmvar' and 'predict.lmvar' support the interval type 'prediction'.
+
+* Input matrices X_mu and X_sigma to 'lmvar' that have only one column, can be of type 'numeric'.
+
+* The matrix of coefficients shown by 'summary', can be restricted to only the coefficients 'beta_mu' or only the coefficients 'beta_sigma'. For this, 'summary.lmvar' has the new arguments 'mu' and 'sigma'. This option is useful when the length of the vector 'beta_mu' and/or 'beta_sigma' is large.
+
+* The function 'df.residuals.lmvar' has been removed. The 'residual degrees of freedom' are a useful concept in a classical linear model where it specifies the degrees of freedom of the Student t-distribution for e.g. confidence intervals of beta. This distribution plays no role in the model of 'lmvar' which uses aymptotically normal distributions.
+
+* Improved documentation, in particular the README and the vignettes 'Intro' and 'Math'
+
+* Bug fixes
+
+Version 1.0.0
+-------------
+
+First release-version.

R/coef.lmvar.R

@@ -20,6 +20,8 @@
 #'
 #' @seealso \code{\link{beta_sigma_names}} for the adaptation of the names of the coefficients in \eqn{\beta_\sigma}.
 #'
+#' \code{\link[stats]{confint}} for the calculation of confidence intervals of \eqn{\beta_\mu} and \eqn{\beta_\sigma}.
+#'
 #' @export
 #'
 #' @example R/examples/coef_examples.R

R/dfree.R

@@ -22,8 +22,6 @@
 #'
 #' @export
 #'
-#' @seealso \code{\link{df.residual.lmvar}} for the residual degrees of freedom for an object of class 'lmvar'
-#'
 #' @example R/examples/dfree_examples.R
 #'
 

R/examples/fitted_examples.R

@@ -32,6 +32,9 @@ fitted(fit)
 # Calculate the expected values, the standard deviations and their 95% confidence intervals
 fitted(fit, interval = "confidence")
 
+# Calculate the expected values and the 90% prediction intervals
+fitted(fit, interval = "prediction", level = 0.9)
+
 # Fit the log of 'accel'
 y = log(attenu$accel)
 fit_log = lmvar(y, X, X_s)
@@ -45,3 +48,5 @@ fitted(fit_log, log = TRUE)
 # Calculate the expected values and the standard deviations of 'accel',
 # as well as their 90% confidence intervals
 fitted(fit_log, log = TRUE, interval = "confidence", level = 0.9)
+
+

R/examples/lmvar_examples.R

@@ -17,16 +17,13 @@ X_s = cbind(attenu$mag, 1 / attenu$dist)
 colnames(X_s) = c("mag", "dist_inv")
 
 # Carry out the fit
-fit_lmvar  = lmvar(attenu$accel, X, X_s)
+fit = lmvar(attenu$accel, X, X_s)
 
-# Inspect the results. Note from the p-value for the difference in
-# deviance that this fit appears to be significantly better than
-# a classical linear fit
-summary(fit_lmvar)
+# Inspect the results
+summary(fit)
 
-# Carry out a classical linear fit for comparison
-fit_lm = lm(attenu$accel ~ mag + dist, attenu)
-
-# A comparison of the AIC values also favours the fit with 'lmvar'
-AIC(fit_lm)
-AIC(fit_lmvar)
+# Carry out the fit with an inital estimate for beta and ask for
+# a report of the solver-routine
+beta_sigma_start = c(-4, 0, 0)
+fit = lmvar(attenu$accel, X, X_s, slvr_options = list(start = beta_sigma_start), slvr_log = TRUE)
+fit$slvr_log

R/examples/predict_examples.R

@@ -40,7 +40,10 @@ predict(fit, XP, XP_s, mu = FALSE)
 predict(fit, XP, XP_s, sigma = FALSE, interval = "confidence")
 
 # Calculate the standard deviations and their 90% confidence intervals
-predict(fit, XP, XP_s, sigma = FALSE, interval = "confidence", level = 0.9)
+predict(fit, XP, XP_s, mu = FALSE, interval = "confidence", level = 0.9)
+
+# Calculate the expected values and the 90% prediction intervals of 'accel'
+predict(fit, XP, XP_s, sigma = FALSE, interval = "prediction", level = 0.9)
 
 # Change the model and fit the log of 'accel'
 y = log(attenu$accel)

R/examples/summary_examples.R

@@ -16,5 +16,11 @@ fit = lmvar(attenu$accel, X, X_s)
 # Print a summary of the fit
 summary(fit)
 
+# Include only the coefficients beta for the expected values
+summary(fit, sigma = FALSE)
+
+# Include only the coefficients beta for the standard deviations
+summary(fit, mu = FALSE)
+
 # Extract the matrix of coefficients from the summary
 coef(summary(fit))

R/fitted.lmvar.R

@@ -1,19 +1,19 @@
 #' @title Fitted values for an 'lmvar' object
 #'
-#' @description Estimators and confidence intervals for the expected values and standard deviations of the response-vector \eqn{Y} of
-#' an 'lmvar' model. The model can be a fit
-#' to the response vector or to the logarithm of the response vector.
+#' @description Estimators and confidence intervals for the expected values and standard deviations of the response vector \eqn{Y} of
+#' an 'lmvar' model. Prediction intervals for \eqn{Y} as well. Alternatively, estimators and intervals can be for \eqn{e^Y}.
 #'
 #' @param object An 'lmvar' object
 #' @param mu Boolean, specifies whether or not to return the expected values
 #' @param sigma Boolean, specifies whether or not to return the standard deviations
-#' @param log Boolean, specifies whether the observations \eqn{Y} have been fitted or the logarithm \eqn{\log Y}. In both cases,
-#' \code{fitted.lmvar} returns expected values and standard deviations for \eqn{Y} itself.
+#' @param log Boolean, specifies whether expected values, standard deviations (as well as their confidence intervals) and
+#' prediction intervals should be for \eqn{Y} (\code{log = FALSE}) or for \eqn{e^Y} (\code{log = TRUE}).
 #' @param interval Character string, specifying the type of interval. Possible values are
 #' \itemize{
-#' \item "none" No interval
+#' \item "none" No interval, this is the default
 #' \item "confidence" Confidence intervals for the expected values (if \code{mu = TRUE}) and the standard deviation
 #' (if \code{sigma = TRUE})
+#' \item "prediction" Prediction intervals for the response vector \eqn{Y} (\code{log = FALSE}) or for \eqn{e^Y} (\code{log = TRUE})
 #' }
 #' @param level Numeric value between 0 and 1, specifying the confidence level
 #' @param ... For compatibility with \code{\link[stats]{fitted}} generic.
@@ -28,10 +28,12 @@
 #' \itemize{
 #' \item \code{mu} Estimators for the expected value \eqn{\mu}
 #' \item \code{sigma} Estimators for the standard deviation \eqn{\sigma}
-#' \item \code{mu_lwr} Lower bound of the interval for \eqn{\mu}
-#' \item \code{mu_upr} Upper bound of the interval for \eqn{\mu}
-#' \item \code{sigma_lwr} Lower bound of the interval for \eqn{\sigma}
-#' \item \code{sigma_upr} Upper bound of the interval for \eqn{\sigma}
+#' \item \code{mu_lwr} Lower bound of the confidence interval for \eqn{\mu}
+#' \item \code{mu_upr} Upper bound of the confidence interval for \eqn{\mu}
+#' \item \code{sigma_lwr} Lower bound of the confidence interval for \eqn{\sigma}
+#' \item \code{sigma_upr} Upper bound of the confidence interval for \eqn{\sigma}
+#' \item \code{lwr} Lower bound of the prediction interval
+#' \item \code{upr} Upper bound of the prediction interval
 #' }
 #'
 #' @export
@@ -51,11 +53,21 @@
 #' @seealso \code{\link{predict.lmvar}} for expected values, standard deviations and intervals for model matrices different from
 #' the ones present in \code{object}.
 #'
+#' \code{\link{coef.lmvar}} and \code{\link[stats]{confint}} for maximum likelihood estimators and confidence intervals for
+#' \eqn{\beta_\mu} and \eqn{\beta_\sigma}.
+#'
 #' @example R/examples/fitted_examples.R
 #'
 
-fitted.lmvar <- function( object, mu = TRUE, sigma = TRUE, log = FALSE, interval = c("none", "confidence"), level = 0.95, ...){
+fitted.lmvar <- function( object, mu = TRUE, sigma = TRUE, log = FALSE, interval = c("none", "confidence", "prediction"), level = 0.95, ...){
 
-  return(predict.lmvar( object, mu = mu, sigma = sigma, log = log, interval = interval, level = level, ...))
+  # Check interval
+  if (missing(interval)){
+    interval = interval[1]
+  }
+  else{
+    interval = match.arg(interval)
+  }
 
+  return(predict.lmvar( object, mu = mu, sigma = sigma, log = log, interval = interval, level = level, ...))
 }