[SciPy-User] overlapping samples t test

[josef.pktd at gmail.com]

On Sat, Nov 9, 2013 at 11:06 PM, Horea Christian <h.chr at mail.ru> wrote:

>  Hey there - relating to my previous question about t-tests. I have (only
> very little) missing data and I was told on CrossValidated (a stack
> exchange for data analysis) in this thread
> http://stats.stackexchange.com/questions/75046/related-sample-t-test-with-missing-valuesthat I should use an overlapping samples t-test. is this in scipy - or is
> there any easy way to get similar functionality? (I could push it upstream
> if you can tell me how to implement it)
>


I've never heard of overlapping samples t-test.

If it mixes independent and paired t-tests, then it might require
additional assumptions that might not be appropriate for a repeated
measures setting, when we have to worry about unobserved differences across
individuals or correlation within individuals.


Josef



>
> Many Thanks,
>
> --
> Horea Christianhttp://chymera.eu
>
>
> _______________________________________________
> SciPy-User mailing list
> SciPy-User at scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-user
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20131110/5dac8351/attachment.html>