[R] qqnorm in R-0.63.3
Peter Dalgaard BSA
p.dalgaard at biostat.ku.dk
Tue Mar 16 14:12:56 CET 1999
Tino Reinhardt <h0444idm at student.hu-berlin.de> writes:
> try x <- table(rnorm(1000) # or similar and
> qqnorm(x,type="s") # in 0.63.2 and 63.3
x <- rnorm(1000)
> Well, the NEWS-file says `qqnorm() uses ppoints() and returns a list
> with x sorted along y.´ and, IMHO, that explains the slopes.
> But is it generally a bad idea to connect the observed points and
> print a nice line instead?
> and if not, could anyone explain why it should be x sorted along y?
I believe the reason given at the time was Splus compatibility. We're
still not compatible when it comes to NA handling, though. S(-plus
3.4) gives NA where the original arguments has NA, R just omits NA
cases. This is a bit silly, because the major use I can think of for
the current convention is to obtain a parallel vector to x with the
"expected normal scores" (or an approximation thereof).
To get a prettier plot, use qqnorm(sort(x),type="s")
Come to thik of it wouldn't it be nice to have a general sort=
O__ ---- Peter Dalgaard Blegdamsvej 3
c/ /'_ --- Dept. of Biostatistics 2200 Cph. N
(*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk) FAX: (+45) 35327907
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
More information about the R-help