Message-ID: <63261.129.177.44.49.1036058218.squirrel@webmail.nhh.no>
Date: 2002-10-31T09:56:58Z
From: Roger Bivand
Subject: Error in Fields TPS function {svd ...} again
In-Reply-To: <5.1.0.14.2.20021030121115.01db3650@38.168.156.2>
> Thanks for all the helpful responses. I include the data file and the
> syntax file for reference. Again, if I use the fields function, as is,
>
It helped to have your example:
> data <- read.table("splus-data-black.dat")
> dim(data)
[1] 606 2
> summary(data)
V1 V2
Min. :0.0000 Min. : 0.0
1st Qu.:0.0000 1st Qu.: 3.0
Median :0.0000 Median : 36.0
Mean :0.3977 Mean : 247.6
3rd Qu.:0.0000 3rd Qu.: 149.9
Max. :7.0000 Max. :4680.0
> table(data[1])
0 1 2 3 4 5 6 7
488 64 24 10 7 9 2 2
Tps(x, Y, ...) expects x to be a matrix, in the geostatistical "world
view" of the locations of the observations in say 2d or 3d:
> Tps(data[,2], data[,1])
Error in svd(tempM) : error 159 in dsvdc
The distribution of your Y makes it look like counts, both x and Y have
smaller numbers of unique values than n=606:
> length(unique(data[,1]))
[1] 8
> length(unique(data[,2]))
[1] 161
> plot(data[,2], data[,1])
shows that the data are rather different from the ones in example(Tps). So
this doesn't look like a typical numerical problem, but Tps being used on
data for which it was not designed, and failing to trap this state.
Roger
I get the message:
>
> Error in svd(tempM) : error 159 in dsvdc
>
> using traceback, I get:
>
> > traceback()
> 4: stop(paste("error ", z$info, " in dsvdc"))
> 3: svd(tempM)
> 2: Krig(x, Y, cov.function = rad.cov, m = m, decomp = decomp, scale.type
> = scale.type,
> outputcall = Tpscall, p = p, ...)
> 1: Tps(bvolcap, bdsm)
> >
>
>
> if I change the occurrence of svd in the fields package to La.svd, I get
> the error message
>
> > bout <- Tps( bvolcap, bdsm)
> Error in "[<-"(*tmp*, (nt + 1):np, (nt + 1):np, value = temp$v) :
> number of items to replace is not a multiple of replacement
> length
> >
>
> traceback() then gives:
>
> > traceback()
> 2: Krig(x, Y, cov.function = rad.cov, m = m, decomp = decomp, scale.type
> = scale.type,
> outputcall = Tpscall, p = p, ...)
> 1: Tps(bvolcap, bdsm)
> >
>
> Sorry to be so useless, I am a R (S) newbie. Any advice is greatly
> appreciated.
--
Roger Bivand
NHH, Breiviksveien 40, N-5045 Bergen, Norway
(travelling but still accessible)
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._