[R] slow computation of mixed ANOVA using aov

[Peter Dalgaard]

Peter Dalgaard <p.dalgaard at biostat.ku.dk> writes:

> - use a version of R compiled with a fast BLAS, on a fast computer
>   with a lot of RAM... (A ~2K square matrix inversion will take a
>   while, but "hours" sounds a bit excessive). 

To wit:

>  system.time(aov(dv~f1*f2*f3+Error(subject/(f1+f3)),data=d))
[1] 582.46   9.31 619.01   0.00   0.00

i.e. about 10 minutes on an Opteron 240 (dual, but it only seemed to
use one cpu for this task) with 1GB.


-- 
   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