Hello,
I know there are many posts covering this topic already but I can't find any that solves my problem:
I have a 2-way anova design: one factor (group) with 2 levels and one factor (genotype) with 3 levels. To save time I'm testing interactions and main effects separately, so I can run them at the same time (just checking - i hope this is ok). I'm testing for 6 interactions, and when testing these normally it all seems fine. The problem occurs when I'm trying to add an F-test for all interactions together.
I get the "singular matrix" error as attached below this email (reported by other people as well).
I suspect this has something to do with my design files, particularly the .fts matrix. So here we go:
My .mat file (assuming 1 subject per genotypeXgroup cell):
/NumWaves 6
/NumPoints 177
/PPheights
/Matrix
1 0 0 0 0 0 (group1 genotype1)
0 1 0 0 0 0 (group2 genotype1)
0 0 1 0 0 0 (group1 genotype2)
0 0 0 1 0 0 (group2 genotype2)
0 0 0 0 1 0 (group 1 genotype3)
0 0 0 0 0 1 (group2 genotype3)
My .con file
/NumWaves 6
/NumContrasts 6
/PPheights
/Matrix
2 -2 -1 1 -1 1
-2 2 1 -1 1 -1
1 -1 1 -1 -2 2
-1 1 -1 1 2 -2
1 -1 0 0 -1 1
-1 1 0 0 1 -1
My .fst file
/NumWaves 6
/NumContrasts 1
/Matrix
1 1 1 1 1 1
Apologies for this long explanation! At least I hope it's kind of clear...
Thanks,
Emma
--------------------------------------------------------------------------------
Data loaded
ERROR: Program failed
An exception has been thrown
Runtime error:- detected by Newmat: matrix is singular
MatrixType = Crout # Rows = 6; # Cols = 6
Trace: Crout(lubksb); GeneralSolvI; InvertedMatrix::Evaluate.
Exiting
----------------------------------------------------------------------------------
|