From 710c849e07d10da7c082ff7352b0740cd20a6e39 Mon Sep 17 00:00:00 2001 From: Japal Date: Wed, 19 Jun 2024 09:28:42 +0200 Subject: [PATCH] Minor fix doc --- man/lrEM.Rd | 4 ++-- man/lrEMplus.Rd | 2 +- zCompositions.Rproj | 4 +--- 3 files changed, 4 insertions(+), 6 deletions(-) diff --git a/man/lrEM.Rd b/man/lrEM.Rd index 9a608f8..f85ee41 100644 --- a/man/lrEM.Rd +++ b/man/lrEM.Rd @@ -37,7 +37,7 @@ If \code{ini.cov="multRepl"}, parameter for initial multiplicative simple replac Maximum number of iterations for the EM algorithm (default = 50). } \item{rlm.maxit}{ -If \code{rob=TRUE}, maximum number of iterations for the embedded robust regression estimation (default = 150; see \code{\link{rlm}} in \code{MASS} package for details). +If \code{rob=TRUE}, maximum number of iterations for the embedded robust regression estimation (default = 150; see \code{\link[MASS]{rlm}} for details). } \item{imp.missing}{If \code{TRUE} then unobserved data identified by \code{label} are treated as missing data (default = \code{FALSE}).} \item{suppress.print}{ @@ -57,7 +57,7 @@ It produces an imputed data set on the same scale as the input data set. If \cod Under maximum likelihood (ML) estimation (default, \code{rob=FALSE}), a correction factor based on the residual covariance obtained by censored regression is applied for the correct estimation of the conditional covariance matrix in the maximisation step of the EM algorithm. This is required in order to obtain the conditional expectation of the sum of cross-products between two components in the case that both involve imputed values. Note that the procedure is based on the oblique additive log-ratio (alr) transformation to simplify calculations and alleviates computational burden. Nonetheless, the same results would be obtained using an isometric log-ratio transformation (ilr). Note also that alr requires at least one complete column. Otherwise, a preliminary imputation, e.g. by \code{\link{multRepl}} or \code{\link{multLN}}, of the most simplest censoring pattern may be enough. The argument \code{ini.cov} determines how the initial estimation of the log-ratio covariance matrix required to start the EM process is worked out. Note that the estimation of the covariance matrix, and hence the lrEM routine, requires a regular data set, i.e. having more observations than variables in the data. -Under robust estimation (\code{rob=TRUE}), the algorithm requires ilr transformations in order to satisfy requirements for robust estimation methods (MM-estimation by default, see \code{rlm} function for more details). An initial estimation of nondetects is required to get the algorithm started. This can be based on either the subset of fully observed cases (\code{ini.cov="complete.obs"}) or a multiplicative simple replacement of all nondetects in the data set (\code{ini.cov="multRepl"}). Note that the robust regression method involved includes random elements which can, occasionally, give rise to \code{\link{NaN}} values getting the routine execution halted. If this happened, we suggest to simply re-run the function once again. +Under robust estimation (\code{rob=TRUE}), the algorithm requires ilr transformations in order to satisfy requirements for robust estimation methods (MM-estimation by default, see \code{\link[MASS]{rlm}} function for more details). An initial estimation of nondetects is required to get the algorithm started. This can be based on either the subset of fully observed cases (\code{ini.cov="complete.obs"}) or a multiplicative simple replacement of all nondetects in the data set (\code{ini.cov="multRepl"}). Note that the robust regression method involved includes random elements which can, occasionally, give rise to \code{\link{NaN}} values getting the routine execution halted. If this happened, we suggest to simply re-run the function once again. Note that conditional imputation based on log-ratio coordinates cannot be conducted when there exist censoring patterns including samples with only one observed component. As a workaround, \code{lrEM} applies multiplicative simple replacement (\code{\link{multRepl}}) on those and a warning message identifying the problematic cases is printed. Alternatively, it might be sensible to simply remove those non-informative samples from the data set. diff --git a/man/lrEMplus.Rd b/man/lrEMplus.Rd index 88881b8..6501e47 100644 --- a/man/lrEMplus.Rd +++ b/man/lrEMplus.Rd @@ -36,7 +36,7 @@ If \code{ini.cov="multRepl"}, parameter for initial multiplicative simple replac Maximum number of iterations (default = 50). } \item{rlm.maxit}{ -If \code{rob=TRUE}, maximum number of iterations for the embedded robust regression estimation (default = 150; see \code{\link{rlm}} in \code{MASS} package for details). +If \code{rob=TRUE}, maximum number of iterations for the embedded robust regression estimation (default = 150; see \code{\link[MASS]{rlm}} for details). } \item{suppress.print}{ Suppress printed feedback (\code{suppress.print = FALSE}, default). diff --git a/zCompositions.Rproj b/zCompositions.Rproj index 5e69648..d1d41d5 100644 --- a/zCompositions.Rproj +++ b/zCompositions.Rproj @@ -10,10 +10,8 @@ NumSpacesForTab: 2 Encoding: UTF-8 RnwWeave: knitr -LaTeX: pdfLaTeX +LaTeX: XeLaTeX BuildType: Package PackageUseDevtools: Yes PackageInstallArgs: --no-multiarch --with-keep.source -PackageCheckArgs: R CMD check --as-cran -PackageRoxygenize: rd,collate,namespace