[R] convolve bug?

[yudi@ucd.ie]

> I am still bugged by the 2nd problem: when I convolve a bump and a step I
> expect an ogive. Instead I get a bump. This seems to be due to a piecing
> together of the expected ogive with its time-reflected self.
> 
You do get what you expect if you shorten the filter and take the piece
of the output that correspond to a full filtering (we lose half of 
the fitler length on both sides.) The zero padding and the filter
length together create the unexpected behavior you mentioned.
 x<-seq(1,500)
 h<-dnorm(c(1:100), mean=50, sd=10)
 y<-ifelse(x<=250, 0, 1)
 plot(convolve(y,h, type="open")[101:500],type='l')   #100= filter length
abline(v=250-50)   # location of original step in the output
points(pnorm(c(1:400),mean=250-50,sd=10))


-Yudi-

------------------------------
Yudi Pawitan: yudi at stat.ucc.ie
Department of Statistics, UCC
Cork, Ireland
Ph : 353-21-902 906
Fax: 353-21-271 040
------------------------------

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._