[pymvpa] Z-scoring raw data

Yaroslav Halchenko debian at onerussian.com
Thu Apr 19 21:15:04 UTC 2012

and depending on how you want to interpret your data z-scoring against
rest condition might have clearer 'meaning' that standartization of the
whole time series happen you to look at the actual values

but, as you have observed, since rest-condition is scarser in # of
samples than the whole time series, its stat estimates are less reliable
thus there could indeed be negative impact on the generalization

On Thu, 19 Apr 2012, Michael Hanke wrote:
> > am I am wondering if you have found it beneficial to z-score in that way or
> > if just doing a simple z-score on the entire time course (per run) provides
> > similar classification accuracy.

> Main reason of doing that in the tutorial was to show that it is
> possible ;-) it wasn't meant to be an advice that it should always be
> done this way.
Keep in touch                                     www.onerussian.com
Yaroslav Halchenko                 www.ohloh.net/accounts/yarikoptic

More information about the Pkg-ExpPsy-PyMVPA mailing list