R help - Aug 2005 - Matrix oriented computing

Hi,

I want to compute the quantiles of Chi^2 distributions with different 
degrees of freedom like

x<-cbind(0.005, 0.010, 0.025, 0.05, 0.1, 0.5, 0.9, 0.95, 0.975, 0.99, 0.995)
df<-rbind(1:100)
m<-qchisq(x,df)

and hoped to get back  a  length(df) times length(x)  matrix with the 
quantiles. Since this does not work, I use

x<-c(0.005, 0.010, 0.025, 0.05, 0.1, 0.5, 0.9, 0.95, 0.975, 0.99, 0.995)
df<-c(1:100)
m<-qchisq(x,df[1])
for (i in 2:length(df)) {
  m<-rbind(m,qchisq(x,df[i]))
}
dim(m)<-c(length(df),length(x))

Is there a way to avoid the for loop ?

Thanks Sigbert

On Fri, 26 Aug 2005 14:44:10 +0200 Sigbert Klinke wrote:
> Hi,
> 
> I want to compute the quantiles of Chi^2 distributions with different 
> degrees of freedom like
> 
> x<-cbind(0.005, 0.010, 0.025, 0.05, 0.1, 0.5, 0.9, 0.95, 0.975, 0.99,
> 0.995) df<-rbind(1:100)
> m<-qchisq(x,df)
> 
> and hoped to get back  a  length(df) times length(x)  matrix with the 
> quantiles. Since this does not work, I use
> 
> x<-c(0.005, 0.010, 0.025, 0.05, 0.1, 0.5, 0.9, 0.95, 0.975, 0.99,
> 0.995) df<-c(1:100)
> m<-qchisq(x,df[1])
> for (i in 2:length(df)) {
>   m<-rbind(m,qchisq(x,df[i]))
> }
> dim(m)<-c(length(df),length(x))
> 
> Is there a way to avoid the for loop ?
You could use sapply():
  m <- sapply(x, function(x) qchisq(x, df))
hth,
Z
> Thanks Sigbert
> 
> ______________________________________________
> 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
>

On Fri, 2005-08-26 at 14:44 +0200, Sigbert Klinke wrote:
> Hi,
> 
> I want to compute the quantiles of Chi^2 distributions with different 
> degrees of freedom like
> 
> x<-cbind(0.005, 0.010, 0.025, 0.05, 0.1, 0.5, 0.9, 0.95, 0.975, 0.99,
0.995)
> df<-rbind(1:100)
> m<-qchisq(x,df)
> 
> and hoped to get back  a  length(df) times length(x)  matrix with the 
> quantiles. Since this does not work, I use
> 
> x<-c(0.005, 0.010, 0.025, 0.05, 0.1, 0.5, 0.9, 0.95, 0.975, 0.99, 0.995)
> df<-c(1:100)
> m<-qchisq(x,df[1])
> for (i in 2:length(df)) {
>   m<-rbind(m,qchisq(x,df[i]))
> }
> dim(m)<-c(length(df),length(x))
> 
> Is there a way to avoid the for loop ?
> 
> Thanks Sigbert
See ?sapply

x <- c(0.005, 0.010, 0.025, 0.05, 0.1, 0.5, 0.9, 
       0.95, 0.975, 0.99, 0.995)

df <- c(1:100)

mat <- sapply(x, qchisq, df)
> dim(mat)
[1] 100  11
 
> str(mat)
 num [1:100, 1:11] 3.93e-05 1.00e-02 7.17e-02 2.07e-01 4.12e-01 ...


HTH,

Marc Schwartz

I believe that the following is what you want:

x <- c(0.005, 0.010, 0.025, 0.05, 0.1, 0.5, 0.9, 0.95, 0.975, 0.99, 0.995)
dof <- 1:100
ans <- outer(x, dof, qchisq)
dimnames(ans) <- list(x, dof)


Note that 'df' is not a very auspicious name for an object since
it is the name of a function. 

Patrick Burns
patrick at burns-stat.com
+44 (0)20 8525 0696
http://www.burns-stat.com
(home of S Poetry and "A Guide for the Unwilling S User")

Sigbert Klinke wrote:
>Hi,
>
>I want to compute the quantiles of Chi^2 distributions with different 
>degrees of freedom like
>
>x<-cbind(0.005, 0.010, 0.025, 0.05, 0.1, 0.5, 0.9, 0.95, 0.975, 0.99,
0.995)
>df<-rbind(1:100)
>m<-qchisq(x,df)
>
>and hoped to get back  a  length(df) times length(x)  matrix with the 
>quantiles. Since this does not work, I use
>
>x<-c(0.005, 0.010, 0.025, 0.05, 0.1, 0.5, 0.9, 0.95, 0.975, 0.99, 0.995)
>df<-c(1:100)
>m<-qchisq(x,df[1])
>for (i in 2:length(df)) {
>  m<-rbind(m,qchisq(x,df[i]))
>}
>dim(m)<-c(length(df),length(x))
>
>Is there a way to avoid the for loop ?
>
>Thanks Sigbert
>
>______________________________________________
>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
>
>
>
>  
>