[R] normal score transform

[Ole Christensen]

Dear Ulrich

One way to back-transform is by simulation.

Procedure 

1. Make 1000 simulation from a normal distribution with mean = predicted
mean, variance = prediction variance
2. back-transform the simulations
3. Calculate mean and variance of the simulations.

For the normal-score-transform you need to define what you mean by the
inverse of such transform 
[as a function taking any real number, not just one of your data
values].
That sounds odd to me, so why not use another transformation ? 

Box-Cox transformations are implemented in package geoR 
[and if I remember correctly, back-transformation by simulations is also
implemented]

Cheers Ole


> But is it also possible to transform the prediction VARIANCE back to non 
> normalized variances? I do not have these variances before hand. Or would be 
> a solution to use the standard deviation as this will be within the range of 
> the predictions? I'm not quite sure... 


-- 
Ole F. Christensen
Department of Mathematics and Statistics
Fylde College, Lancaster University 
Lancaster, LA1 4YF, England
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._