[Rd] nlme package. Minor change in method print.VarCov (PR#10669)

agalecki at umich.edu agalecki at umich.edu
Thu Jan 31 17:55:07 CET 2008


Hello,

I would like to propose to make a minor change in  method  print.VarCov 
by inserting one line
            stdevs <-  FALSE                          #  Inserted line
as shown later in this email.

By adding this line we would modify the output from
print(corrmtx) as shown in the illustration below. More specifically the
line with redundant information, which reads  " Standard Deviations: 1 1 1 "
would be skipped from the output.

Otherwise, to omit line with SDs:1 1 1 from the output  we need to use 
more cumbersome syntax:
print(corrmtx, corr=TRUE, stdevs=FALSE).

Output from sessionInfo()  included at the bottom of this email

Sincerely,

Andrzej Galecki



============ Illustration  ========
 > selected.rows <- seq(1,308, by=10)
 > ovary <- Ovary[selected.rows,]       # selected rows
 > fm1 <- gls(follicles ~ sin(2*pi*Time) + cos(2*pi*Time), ovary, 
correlation = corAR1(form = ~ 1 | Mare))
 > vmtx <- getVarCov(fm1)   # Covariance matrix
 > corrmtx <- cov2cor(vmtx)  # Correlation matrix
 > print(corrmtx, corr=TRUE)
Marginal correlation matrix
        [,1]    [,2]    [,3]
[1,] 1.00000 0.39172 0.15345
[2,] 0.39172 1.00000 0.39172
[3,] 0.15345 0.39172 1.00000
  Standard Deviations: 1 1 1
 > print(corrmtx,corr=TRUE,stdevs=FALSE)
Marginal correlation matrix
        [,1]    [,2]    [,3]
[1,] 1.00000 0.39172 0.15345
[2,] 0.39172 1.00000 0.39172
[3,] 0.15345 0.39172 1.00000



==========Part of the Source code with suggested change  =======

 > getS3method("print","VarCov")
function (x, corr = FALSE, stdevs = TRUE, digits = 5, ...)
{
    pvc <- function(x, type, corr, stdevs, digits) {
        cat(c("Random effects", "Conditional", "Marginal")[match(type,
            c("random.effects", "conditional", "marginal"))],
            " ", sep = "")
        x <- as.matrix(x)
        class(x) <- NULL
        attr(x, "group.levels") <- NULL
        if (corr) {
            cat("correlation matrix\n")
            sds <- sqrt(diag(x))
            print(signif(t(x/sds)/sds, digits))
            stdevs <-  FALSE                          #  Inserted line
       }
        else {
            cat("variance covariance matrix\n")
            print(signif(x, digits))
            if (stdevs)
                sds <- sqrt(diag(x))
        }
 


=======================================================================

 > sessionInfo()
R version 2.6.1 (2007-11-26)
i386-pc-mingw32

locale:
LC_COLLATE=English_United States.1252;LC_CTYPE=English_United 
States.1252;LC_MONETARY=English_United 
States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] nlme_3.1-86

loaded via a namespace (and not attached):
[1] grid_2.6.1     lattice_0.17-2 tools_2.6.1   
 >



More information about the R-devel mailing list