[R] Question about the smooth.Pspline

[Petr Klasterecky]

Xuhong Zhu napsal(a):
> Hello, Everyone,
> 
> I want to use the smooth.Pspline to smooth my data but R give me the
> error message as follows:
> 
> Error in smooth.Pspline(sort.e$time, sort.e$cuff, method = 3) :
>         X not strictly increasing
> 
> my data looks like the following:
> 
> id       cuff           time     patient
> ...
> 2783 13.229608  478       6
> 3472 20.904825  478       7
> 4155 15.033727  478       8
> 4845 19.342963  478       9
> 715   8.000000  479       3
> 1422 22.052385  479       4
> 2110 15.393063  479       5
> 2784 13.200922  479       6
> 3473 20.900132  479       7
> ...
> 
> my R codes is:
> 
> e <- rbind(patient.1,patient.2,patient.3,...)
> attach(e)
> sort.e <- e[order(time),]
> plot(sort.e$time, sort.e$cuff, xlab="Time", ylab="Cuff",type="p", col=3,
> xlim=c(c[2],d[2]) , ylim=c(c[1], d[1]) , main="one Smooth Curve for 10
> Patients")
> fm <- smooth.Pspline(sort.e$time, sort.e$cuff, method=3)
> lines(fm$x, fm$y, lty=1,col=1)
> 
> 
> What I am doing here is to combine the data together and find a smooth
> curve. My question is if the smooth.Pspline could not be used in my
> data since the variable "time" has repeated values.

Quoted from your error message:
 > Error in smooth.Pspline(sort.e$time, sort.e$cuff, method = 3) :
 >         X not strictly increasing

Quoted from help of smooth.Pspline (which, as you should have told us by 
the way, is contained in package pspline):

Arguments:
x 	values of the predictor variable. These must be strictly increasing, 
and there must be at least 2*norder + 1 of them.
sm.spline provides a simplified interface, in which the x values can be 
unsorted, or a list with components "x" and "y" or a two-column matrix 
or a complex vector.

Strictly increasing = no repeated values...
Petr

> 
> Xuhong
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 

-- 
Petr Klasterecky
Dept. of Probability and Statistics
Charles University in Prague
Czech Republic