Dear Dr. Maris,<br><br>thanks for your reply.<br>If I get you
right you say that whether I use the statfun_depsamplesregrT or
use/write statfun_depsamplesrankcorr depends on assumed (apriori)
relation between the predictor variable (experimental condition) and the
data. If I assume linearity (i.e. in the case of WM load: ERP/TF(1)
< ERP/TF(2) < ERP/TF(3) < ERP/TF(4)) I might use the
statfun_depsamplesregrT. If I doubt linearity (i.e.WM load ERP/TF(1)
< ERP/TF(2) <= ERP/TF(3) < ERP/TF(4)) I should use the
statfun_depsamplesrankcorr. Is this correct?<br>
<br>I am not sure if I underestood your point about reference
distribution. Are you saying that for the cluster-based permutation
inference I need to threshold on a reference distribution which might be
either parametric or non-parametric? If this is the case. What does it
depend on whether I use parametric or non parametric reference
distribution?<br>
<br>Thank you again for your time and your help.<br><br>Best,<br>Marcin<br><br><div class="gmail_quote">W dniu 22 marca 2012 21:12 użytkownik Eric Maris <span dir="ltr"><<a href="mailto:e.maris@psych.ru.nl">e.maris@psych.ru.nl</a>></span> napisał:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div link="blue" vlink="purple" lang="EN-US"><div><p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d">Dear Marcin,<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d"><u></u> <u></u></span></p><p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d">The statfun_depsamplesregrT calculates a T-statistic for regression coefficients that are calculated within each of the units-of-observation (typically, participants) obtained by regressing the subject-specific data (spatiotemporal, spatio-spectral, spatio-spectro-temporal) on some predictor variable that varies over the different conditions in which this participant has provided data (e.g., working-memory load, retention interval, luminance, contrast, etc). If you doubt the assumed linear relation between predictor variable and biological data, then you could write your own statfun_depsamplesrankcorr. To use this test statistic for cluster-based permutation inference, you need a threshold based on some reference distribution (which can be parametric, but must not be). <u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d"><u></u> <u></u></span></p><p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d">To get this statfun_depsamplesrankcorr running, you will probably have to take a look in the Fieldtrip code to see how the statistics framework is structured.<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d"><u></u> <u></u></span></p><p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d">Best,<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d"><u></u> <u></u></span></p><p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d">Eric Maris<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d"><u></u> <u></u></span></p><p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Verdana","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<div style="border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt"><div><div style="border:none;border-top:solid #b5c4df 1.0pt;padding:3.0pt 0cm 0cm 0cm"><p class="MsoNormal"><b><span style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">From:</span></b><span style="font-size:10.0pt;font-family:"Tahoma","sans-serif""> Marcin [mailto:<a href="mailto:m.leszczynski.m@googlemail.com" target="_blank">m.leszczynski.m@googlemail.com</a>] <br>
<b>Sent:</b> woensdag 21 maart 2012 10:51<br><b>To:</b> Email discussion list for the FieldTrip project<br><b>Subject:</b> [FieldTrip] statfun_depsamplesregrT<u></u><u></u></span></p></div></div><div><div class="h5"><p class="MsoNormal">
<u></u> <u></u></p><p class="MsoNormal">Dear Fieldtripers,<br><br>Could anyone explain me what is being calculated with the statfun_depsamplesregrT function, please.<br><br>David Groppe (thank you David) suggested in a previous thread on the list that I might calculate permutation test based on rank correlation to account for monotonic relationships within the permutation framework. I was wondering if this is the kind of test that statfun_depsamplesregrT function calculates.<br>
<a href="http://mailman.science.ru.nl/pipermail/fieldtrip/2011-December/004578.html" target="_blank">http://mailman.science.ru.nl/pipermail/fieldtrip/2011-December/004578.html</a><br><br>Best,<br>Marcin<u></u><u></u></p></div>
</div></div></div></div><br>_______________________________________________<br>
fieldtrip mailing list<br>
<a href="mailto:fieldtrip@donders.ru.nl">fieldtrip@donders.ru.nl</a><br>
<a href="http://mailman.science.ru.nl/mailman/listinfo/fieldtrip" target="_blank">http://mailman.science.ru.nl/mailman/listinfo/fieldtrip</a><br></blockquote></div><br>