<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns="http://www.w3.org/TR/REC-html40"
xmlns:ns0="http://schemas.microsoft.com/office/2004/12/omml">

<head>
<meta http-equiv=Content-Type content="text/html; charset=us-ascii">
<meta name=Generator content="Microsoft Word 11 (filtered medium)">
<style>
<!--a:link
        {mso-style-priority:99;}
span.MSOHYPERLINK
        {mso-style-priority:99;}
a:visited
        {mso-style-priority:99;}
span.MSOHYPERLINKFOLLOWED
        {mso-style-priority:99;}

 /* Font Definitions */
 @font-face
        {font-family:Verdana;
        panose-1:2 11 6 4 3 5 4 4 2 4;}
 /* Style Definitions */
 p.MsoNormal, li.MsoNormal, div.MsoNormal
        {margin:0cm;
        margin-bottom:.0001pt;
        font-size:12.0pt;
        font-family:"Times New Roman";}
a:link, span.MsoHyperlink
        {color:blue;
        text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
        {color:purple;
        text-decoration:underline;}
p.formatvorlagearial10ptvor5ptnach5pt, li.formatvorlagearial10ptvor5ptnach5pt, div.formatvorlagearial10ptvor5ptnach5pt
        {margin:0cm;
        margin-bottom:.0001pt;
        font-size:10.0pt;
        font-family:Arial;}
p.formatvorlagearial10ptvor5ptnach5pt1, li.formatvorlagearial10ptvor5ptnach5pt1, div.formatvorlagearial10ptvor5ptnach5pt1
        {margin:0cm;
        margin-bottom:.0001pt;
        font-size:10.0pt;
        font-family:Arial;}
span.E-MailFormatvorlage19
        {mso-style-type:personal;
        font-family:Arial;
        color:windowtext;}
span.E-MailFormatvorlage20
        {mso-style-type:personal;
        font-family:Verdana;
        color:#1F497D;}
span.E-MailFormatvorlage21
        {mso-style-type:personal;
        font-family:Verdana;
        color:#1F497D;}
span.E-MailFormatvorlage22
        {mso-style-type:personal-reply;
        font-family:Arial;
        color:navy;}
@page Section1
        {size:612.0pt 792.0pt;
        margin:72.0pt 90.0pt 72.0pt 90.0pt;}
div.Section1
        {page:Section1;}
-->
</style>
<!--[if gte mso 9]><xml>
 <o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
 <o:shapelayout v:ext="edit">
  <o:idmap v:ext="edit" data="1" />
 </o:shapelayout></xml><![endif]-->
</head>

<body lang=DE link=blue vlink=purple>

<div class=Section1>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-GB
style='font-size:10.0pt;font-family:Arial;color:navy'>Hi Eric, <o:p></o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-GB
style='font-size:10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-GB
style='font-size:10.0pt;font-family:Arial;color:navy'>thank you very much for
your reply!<o:p></o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-GB
style='font-size:10.0pt;font-family:Arial;color:navy'>I still have a question
regarding your answer, though.<o:p></o:p></span></font></p>

<p class=MsoNormal><font size=2 color="#1f497d" face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:#1F497D'><o:p> </o:p></span></font></p>

<p class=MsoNormal><font size=2 color="#1f497d" face=Verdana><span lang=EN-US
style='font-size:11.0pt;font-family:Verdana;color:#1F497D'>I propose that you
calculate your p-values always one-sided. In fact, this is what the FT
permutation statistics functions also do. The difference between a one- and a
two-sided test is that you compare this one-sided p-value either with your
desired type-I error level (for a one-sided test) or with half your desired
type-I error level (for a two-sided test). <o:p></o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p>

<p class=MsoNormal><font size=2 color="#1f497d" face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:#1F497D'><o:p> </o:p></span></font></p>

<p class=MsoNormal><font size=2 color="#1f497d" face=Verdana><span lang=EN-US
style='font-size:11.0pt;font-family:Verdana;color:#1F497D'>I think that the
difference between 0.036 and 0.056 is due to the fact that these are random
quantities. If expect that, if you would increase cfg.numrandomization to
100,000, you would find two p-values that are much closer. In any case, the
p-values are one-sided, and their calculation is independent of the value that
you choose for cfg.alpha.<o:p></o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:navy'>I understand that
cfg.alpha only gives the threshold for which of my p-values is counted as being
significant. However, I thought, changing cfg.tail would change whether I am looking
for positive and negative clusters (cfg.tail = 0, two-sided test), or only
positive or only negative clusters (cfg.tail = 1 or -1, one-sided test). <o:p></o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:navy'>The change in stat.prob
from 0.036 to 0.056 described in my previous email seems to depend on whether I
set cfg.tail to 0 or 1. I’ve computed it multiple times and witch
cfg.tail = 0, stat.prob evolves around 0.03 and with cfg.tail = 1 around 0.05.<o:p></o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:navy'>Could you maybe explain
this to me? <o:p></o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:navy'>From your explanation
above I would assume that I can use cfg.tail = 0 and cfg.alpha = 0.05 to end up
with a mask including my significant clusters in a one-sided test?<o:p></o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:navy'>Thanks again for your
help!<o:p></o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p>

<p class=MsoNormal><font size=2 color=navy face=Arial><span lang=EN-US
style='font-size:10.0pt;font-family:Arial;color:navy'>Nina<o:p></o:p></span></font></p>

</div>

</body>

</html>