Lmu: Estimate predicted response for a factor level.

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

Usage

Lmu(fit, nam, verbose = 0)

Arguments

fit
nam
verbose

Examples

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

## The function is currently defined as
function (fit, nam, verbose = 0) 
{
    if (class(fit) != "lmer") 
        stop("only implemented for lmer")
    v <- fit@frame[[nam]]
    if (!is.factor(v)) 
        stop("nam needs to specify the name of a factor")
    levs <- levels(v)
    if (verbose > 0) 
        print(levs)
    cmat <- contrasts(v)
    if (verbose > 0) 
        print(cmat)
    fe <- getFix(fit)$fixed
    if (verbose > 0) 
        print(fe)
    if (substring(nam, 1, 1) != "^") 
        nam <- paste("^", nam, sep = "")
    L.indices <- grep(nam, names(fe))
    if (verbose > 0) 
        print(L.indices)
    L <- matrix(0, nrow = length(levs), ncol = length(fe))
    colnames(L) <- names(fe)
    if (verbose > 0) 
        print(L)
    rownames(L) <- levs
    L[, L.indices] <- cmat
    if ("(Intercept)" %in% colnames(L)) 
        L[, "(Intercept)"] <- 1
    L
  }

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