<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-15">
</head>
<body text="#000000" bgcolor="#ffffff">
<font color="#000090"><font size="2"><font face="Arial"><span
style="font-size: 10pt;"><b>A tutti gli interessati, <br>
si segnala il seminario<br>
</b></span></font></font></font><font size="2"><font
face="Arial"><span style="font-size: 10pt;"><br>
<font color="#800000"><b>Beyond traditional forensic cases <br>
</b></font> <br>
<font color="#000090">Prof. Thore Egeland<br>
<i>University of Oslo, Norway<br>
</i></font> <br>
<font color="#000090"><b>giovedì 18 novembre 2010 ore 16.30
aula 24 (3° piano)<br>
Dipartimento di Economia – Via Silvio D’Amico 77, Roma<br>
</b> <br>
<b>Abstract<br>
</b><i>The talk is motivated by a paper by Homer et al.
(PLoS Genetics 2008) which has attracted great attention
also among statisticians. The paper caused data to be
removed from the public domain as it was claimed that the
identity of contributors could be revealed based on the
methods suggested.<br>
The statistical methods of the paper are discussed. I also
show how conventional statistical methods like linear
regression can alternatively be used to analyse mixtures
with a large number of contributors. Examples based on a
controlled experiment will be presented. <br>
</i><b> <br>
</b></font></span></font></font>
<pre class="moz-signature" cols="72">--
Julia Mortera
Dipartimento di Economia
Università Roma Tre
Via Silvio D'Amico 77
00145 Roma
Tel: +39 06 57335732
Fax: +39 06 57335771
URL: <a class="moz-txt-link-freetext" href="http://host.uniroma3.it/dipartimenti/economia/it/index.asp?se=Personale&id=49&incd=docenti">http://host.uniroma3.it/dipartimenti/economia/it/index.asp?se=Personale&id=49&incd=docenti</a>
</pre>
</body>
</html>