Ldiff: Hypothesis matrix for ...

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

Usage

Ldiff(fit, pat, levnames = c(reflevel, substring(rownames(L), cut + 1)), reflevel = "<ref>", cut = nchar(pat), verbose = F)

Arguments

fit
pat
levnames
reflevel
cut
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, pat, levnames = c(reflevel, substring(rownames(L), 
    cut + 1)), reflevel = "<ref>", cut = nchar(pat), verbose = F) 
{
    L <- Lmat(fit, pat)
    nam <- rownames(L)
    n <- nrow(L)
    zm <- matrix(1:n, nrow = n, ncol = n)
    plus <- zm[col(zm) < row(zm)]
    minus <- rep(1:(n - 1), (n - 1):1)
    Lp <- L[plus, ]
    Lm <- L[minus, ]
    Lret <- rbind(L, Lp - Lm)
    pnames <- levnames[c(1:n, plus) + 1]
    mnames <- levnames[c(rep(0, n), minus) + 1]
    if (verbose) {
        print(levnames)
        print(plus)
        print(minus)
        print(Lp)
        print(Lm)
        print(L)
        print(Lret)
        print(pnames)
        print(mnames)
    }
    rn <- paste(levnames[c(1:n, plus) + 1], levnames[c(rep(0, 
        n), minus) + 1], sep = " - ")
    rownames(Lret) <- rn
    Lret
  }

gmonette/spida documentation built on May 17, 2019, 7:25 a.m.