pacf.mts
Ok, I can use acf(as.ts(z), plot=FALSE, type= "partial") whether we call it a work around or the proper usage. (My way has worked for a long time in both R and Splus, but I'm prepared to modernize.) But shouldn't the error message say something about not finding a default method? Paul
Peter Dalgaard BSA wrote:
Paul Gilbert <pgilbert@bank-banque-canada.ca> writes:
I think this is a bug. Sorry I'm still using a version from a couple of days ago.
It is, but maybe not the easiest one to fix, and we're already in code freeze...
library(ts) z <- matrix (rnorm(200), 100,2) zz <- acf(z, plot=FALSE, type= "partial")$acf
Error in eval(expr, envir, enclos) : couldn't find function "pacf.mts"
The issue is that certain programming techniques are not compatible
with namespaces. In this case we have in pacf.default:
if (is.matrix(x)) {
m <- match.call()
m[[1]] <- as.name("pacf.mts")
return(eval(m, parent.frame()))
}
but pacf.mts is not visible in the parent.frame(). Since the content
of the if() block is obviously nonfunctional, I suppose we could
accept a fix that is clearly local to the block. Does it work to set
m[[1]] <- pacf.mts
??
Then again, there's a fairly obvious workaround:
acf(as.ts(z), plot=FALSE, type= "partial")
********
(obviously, the same issues apply to pacf())
--
O__ ---- Peter Dalgaard Blegdamsvej 3
c/ /'_ --- Dept. of Biostatistics 2200 Cph. N
(*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk) FAX: (+45) 35327907
______________________________________________ R-devel@stat.math.ethz.ch mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-devel