xlevels: Create 'derivatives' and 'means' of factors to generate, for...

In gmonette/spida15: Collection of miscellaneous functions for mixed models etc. prepared for SPIDA 2009+ (development version)

Description

The functions xlevels and dlevels are primarily intended to create arguments for expand.grid to create 'prediction' or 'effect data frames', to generate wald tests and estimates of specific effects in models with interactions,

Usage

xlevels(f, type = c("raw", "(mean)"), all = FALSE, sep = "")

dlevels(f, type = "pairwise", all = FALSE, sep = "")

Pmat(f, type = c("factor", "raw", "mean", "(mean)", "<mean>", "II", "cen", "cent", "center", "centre", "<centre>", "<center>", "(center)", "(centre)", "III", "pairwise", "<others.m>", "<others.c>", "diff", "diffmean", "diffc", "diffcen", "diffcentre", "diffcentre"), all = FALSE, sep = "")

Pmat_diffmat(w)

Arguments

f	a factor or otherwise a vector interpreted as the levels of a factor
type	one or more types of contrasts ('derivative') or means (convex combinations) of factor levels
all	for some values of type, indicates whether to use all contrasts. e.g. type = "pairwise" will produce pairwise comparisons in both directions if all == TRUE
sep	can add spaces in constructed factor levels
w	weights used for each factor level in creating contrasts differentiating a factor level from others. Weights corresponding to frequencies in the data frame result in effects corresponding to type II effects while equal weights correspond to type III effects for interacting specific effects

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

gmonette/spida15 documentation built on May 17, 2019, 7:26 a.m.