summary.betaTest andprint.summary.betaTest, to correct screen printing for oneof the objects returned bysummary.betaTest.print.summary.betaTest, to complementsummary.betaTest.summary.betaTest, to provide invisible output,so that the summary table could be extracted without screenprinting..lm.rrpp for mismatched numbersof subjects and data.kcomp, which performs a K-componentanalysis.betaTest, which performs tests ofcoefficients vectors, with enhanced flexibility compared tocoef.lm.rrpp.mahal_dist, which emulates thedist function but allows a covariance matrix to be used forgeneralized distances.pairwise suite of functions:stdist andmdist, for standardized andMalahanobis distances, respectively.groups.first, tomeasurement.error for ordering terms having subjects andgroups in ANOVA.anc.BM.measurement.error.getLSmeans.QRforX for determining 0values.Pcov issue inmanova.update.QRforX, which produces consistent QRdecomposition results despite differences betweenbase::qrandMatrix::qr functions.removeRedundant andgetRank to properly remove axes based on abase::qr pivot strategy, but still avoiding the making oflarge, dense matrices before usingbase::qr.R-dependency to >= 4.4.0checkers function thatassumed sparse-matrix (Matrix) format of an object withoutcoercing this format.predict.lm.rrpp and made the code more universal fordifferent model formulae..coef.lm.rrpp..lm.rrpp associatedwith covariance matrices (causing non-adjusted matrices).subTest output fromlm.rrpp.ws.anova.multi.model andlogL related to missingPcov when GLS modelsare used.lm.rrppandpredict.lm.rrpp, and permutation of full modelresiduals (with restrictions) tolm.rrpp.measurement.error.lm.rrpp.ws.verbose arguments to most analytical functions toreduce dense results.getANOVAStats,getPermInfo,getTerms, andgetModels) to accommodateverbose functionchoices, plus allow users to easily extract objects as output.ICCstats, for calculating ICCstatistics forlm.rrpp fits.measurement.error, butalso added a plot utility function,focusMEonSubjects,which builds off ofplot.measurement.error, to focusattention on specific plot details.interSubVar, a utility function to usealong withmeasurement.error, to visualize how variationbetween replicate measures, within subjects, can affect inter-subjectvariation.summary.lm.rrpp that producesmatrices of 0s. The issue was inherent tolm.rrpp, fromerroneous assignment of models, only for output (not formechanics).predict.lm.rrpp for QR-truncateddesign matrices.print.progess conditions inpairwise.Matrix::qrQ,using rather,Matrix::qr.Q.pairwise.SE added tocoef.lm.rrpp forcalculating bootstrap standard error.plot.lm.rrppplot.ordinate to better work withplot.default.plot.lm.rrpp diagnosticplots.logLik.lm.rrpp to obtainlog-likelihood from anlm.rrpp object.scaleCov to scale covariance matriceswith linear or exponential scalars.model.comparison to include Z scores calculatedfrom log-likelihoods.predict.lm.rrpp for new data frames withonly one observation.SS.iter.main that incidentally wrappedRSS.model by rows rather than columns.summary.lm.rrpp that did not properlyindex a matrix ofRSS.model.plot.lm.rrpp for diagnostic plots,which forced an error.Matrix classmatrices for more efficient computation, when needed.checkers function to use better algorithms toswitch among different class matrices, to better save memory andincrease computation time efficiency.anc.BM for singleton nodes.lm.rrpp to have less detritus during use. Alsoadjusted/updated support functions to work with updates.data.frame objects.lm.args.from.formula for intercept onlymodels with covariance matrices.lm.rrpp support functions.logL support function for non-full rankdesign matrices.na.omit.rrpp.data.frame added for handling missingdata.looCV function added as diagnostic tool.coef.lm.rrpp updated to provide results based onSS.type, rather than type III SS, only.R.coef.lm.rrpplm.rrpp (passed ontolm).as.matrix names dropping in supportcode.lm.rrpp usingMatrix package and sparse matrix calculations. This speedsup computation time and requires far less memory allocation.lm.rrpp function now has a an argument,turbo, which can suppress calculating coefficients inrandom permutations, if unneeded, which can speed up analysis of largedata sets.convert2ggplot, for coercingRRPP plots into ggplot objects.plot.predict.lm.rrpp.effect.size.plot.ordinate.predict.lm.rrpp so that functions in formulaeare permissible.summary.pairwise to perform degreetransformations rather thanprint.summary.pairwise, so thatobjects saved are the same as objects printed.model.comparison.manova.update pc dimension issue (output)xlim andylim to be adjustable inplot.ordinate.manova.update to have more efficient code andbetter notes.summary.manova.lm.rrpp that mixed uprows and columns of a matrix of random stats.lm.rrpp datawhen converting a vector of data to a matrix.rrpp.data.frame to prevent downstream issues.trajectory.analysis withunivariate response data, omitting vector correlations output.ordinate function.summary.ordinate andplot.ordinate S3functionsadd.tree function (for plotting withplot.ordinate)lm.rrpp to provide betterflexibility for different formulas.$LM$data inlm.rrpp to be a modelframe rather than a data frame, consistent with$model fromlm.prep.lda A new function to generate arguments forlda in theMASS library.classify deprecated (in favor ofprep.lda)verbose option withmanova.update. The function was optimized to provideverbose output without having to slow down computation time.SS.iter (produced incorrect RSS).trajectoty.analysis.lm.rrpp function.tol andpc.no arguments tomodel.comparison (were fixed before) so that users havemore control of the analysis.prcomp.lm.rrpp to work better withmissing data frames.trajectory.analysis traj.list issue, to not usegrep for sorting trajectories. (Now lexical ordering of interactions isused.)det todeterminant in all needingfunctions, to use modulus for near-singular matricesplot.lm.rrpp (code lines out oforder)model.comparisonplot.procD.lmmanova.update functiontrajectory.analysis functionreveal.model.designs functionANOVA versus MANOVA in RRPPpairwisefunction.procD.lm to better work with data in the globalenvironment rather than a data frame.print.summary.pairwise.model.comparison function.classify.pairwise.anova.lm.rrpp, related to GLS permutations and interceptonly models.anova.lm.rrpp, so that it can be called by otherfunctions/packages.pairwise function: allows pairwise comparison of meansor slopes for alm.rrpp fit.anova.lm.rrppcoef.lm.rrpp tests when type II or typeIII SS is chosen, to make sure that appropriate coefficients areused.anova.lm.rrpp.r
coef.lm.rrpp.r
lm.rrpp.r
predict.lm.rrpp.r
RRPP.support.code.r
RRPP.utils.r
Added aNEWS.md file to track changes to thepackage.