[FieldTrip] Error in correlation analysis with ft_freqstatistics

[jan-mathijs schoffelen]

Dear Rodolphe,

The trick here is not to specify cfg.statistic = 'indepsamplesF';
You need to specify cfg.statistic = 'yourownfunction', and create an m- 
file 'statfun_yourownfunction.m' that computes a correlation. That  
could look like

function s = statfun_yourownfunction(cfg, dat, design)

% mean subtract
mdat = mean(dat,2);
dat = dat - mdat(:,ones(1,size(dat,2)));

mdesign = mean(design);
design = design - mean(design);

% covariance
covdat = dat*design';

% variance
vdat = sum(dat.^2,2);
vdes = sum(design.^2);

% correlation
s.stat = covdat./sqrt(vdat.*vdes);

This is just typed down in half a minute so don't take it for granted.

Best,

Jan-Mathijs


On Jan 25, 2011, at 9:02 PM, Rodolphe Nenert wrote:

> Dear Fieldtrip Users,
>
> im actually trying to do a correlation analysis between frequency  
> power and behavioral results.
> The problem is when i change the design matrix containing the  
> classical independent variables to behavioral scores, as indicate in  
> the walktrough section of the fieldtrip website :
>
> cfg.design = [ .1 .2 .4 .3 etc...];
> cfg.ivar = 1;
>
> Then, Matlab gives me the following errors:
>
> ??? Error using ==> statfun_indepsamplesF at 55
> Invalid specification of the independent variable in the design array.
>
> Error in ==> statistics_montecarlo at 241
>   [statobs, cfg] = statfun(cfg, dat, design);
>
> Error in ==> statistics_wrapper at 285
>     [stat, cfg] = statmethod(cfg, dat, cfg.design,  
> 'issource',issource);
>
> Error in ==> ft_freqstatistics at 127
>   [stat, cfg] = statistics_wrapper(cfg, varargin{:});
>
> Does anyone have an example of a script for such analysis that works?
>
> Thanks,
>
> Rodolphe.
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip

Dr. J.M. (Jan-Mathijs) Schoffelen
Donders Institute for Brain, Cognition and Behaviour,
Centre for Cognitive Neuroimaging,
Radboud University Nijmegen, The Netherlands
J.Schoffelen at donders.ru.nl
Telephone: 0031-24-3614793