<div dir="ltr"><div class="gmail_quote"><div style="font-size:10pt;font-family:Verdana,Geneva,sans-serif"><p><br><br></p>
<p style="text-align:center"><img src="cid:15245954415adf7af1515a9333203227@unb.br" width="597" height="51"></p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"><span style="font-size:14pt"><b>O Programa de Pós-Graduação em Estatística convida para:</b></span></p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"> </p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"> </p>
<p class="m_3898422994082169068western" style="text-align:center"><span style="color:#ffffff;font-size:18pt;background-color:#000000"><b>PALESTRA:</b></span></p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"> </p>
<p style="text-align:center"> </p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"><span style="font-family:'book antiqua',palatino,serif;font-size:24pt"><b><span style="color:#000000"><u>“</u></span></b><b><a class="m_3898422994082169068western" href="http://www.pgest.unb.br/seminarios/163-functional-regression-approximate-bayesian-computation-for-gaussian-process-density-estimation" target="_blank"><span style="color:#000000"><span><span><u>Functional regression approximate Bayesian computation for Gaussian process density estimation</u></span></span></span></a></b><b><span style="color:#000000"><span><u>”</u></span></span></b></span></p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"> </p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"><span style="font-size:xx-large"><b>Palestrante: </b></span></p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"><span style="font-family:'Times New Roman',serif"><span style="font-size:xx-large">Prof. Guilherme Souza Rodrigues (EST/UnB)</span></span></p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"> </p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"> </p>
<p class="m_3898422994082169068western" style="text-align:center"><span style="font-size:14pt;font-family:'book antiqua',palatino,serif"><span><b>DATA: </b></span><span style="color:#000000"><span>26/04</span></span><span style="color:#000000"><span>/2018</span></span><span style="color:#000000"><span> </span></span><span>(</span><span style="color:#000000"><span><span>quinta</span></span></span><span>-feira)</span></span></p>
<p class="m_3898422994082169068western" style="text-align:center"><span style="font-size:14pt;font-family:'book antiqua',palatino,serif"><span><b>HORÁRIO: </b></span><span>14:00h </span></span></p>
<p class="m_3898422994082169068western" style="text-align:center"><span style="font-size:14pt;font-family:'book antiqua',palatino,serif"><span><b>LOCAL: </b></span><span style="color:#000000"><span>Sala Multiuso EST (A1-7/76)</span></span></span><span style="color:#000000"><span style="font-size:xx-large"><span> </span></span></span></p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"> </p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"><span style="font-family:'Times New Roman',serif"><span style="font-size:xx-large"><b>Resumo</b></span></span></p>
<p class="m_3898422994082169068western" style="text-align:center" align="center"> </p>
<p class="m_3898422994082169068western" style="text-align:justify" align="justify"><span style="font-family:'Times New Roman',serif;font-size:14pt"><span><span style="color:#777777"><span style="font-family:Verdana,Geneva,sans-serif"><span>A</span></span></span> novel Bayesian nonparametric method is proposed for hierarchical modelling on a set of related density functions, where grouped data in the form of samples from each density function are available. Borrowing strength across the groups is a major challenge in this context. To address this problem, a hierarchically structured prior, defined over a set of univariate density functions using convenient transformations of Gaussian processes, is introduced. Inference is performed through approximate Bayesian computation (ABC) via a novel functional regression adjustment. The performance of the proposed method is illustrated via simulation studies and an analysis of rural high school exam performance in Brazil. </span></span></p>
<p style="text-align:center"> </p>
<p style="text-align:center"><span style="color:#000000;background-color:#99ccff;font-family:'arial black',sans-serif;font-size:12pt"><b> VEJA OS PRÓXIMOS SEMINÁRIOS: <a style="background-color:#99ccff;color:#000000" href="http://www.pgest.unb.br/seminarios" rel="noreferrer" target="_blank">http://www.pgest.<wbr>unb.br/seminarios</a> </b></span></p>
<p style="text-align:center"> </p>
<div style="text-align:center">______________________________<wbr>______________________________<wbr>______________________________<wbr>________________<span class="HOEnZb"><font color="#888888"><br>
<div class="m_3898422994082169068pre" style="margin:0;padding:0;font-family:monospace"> </div>
<div class="m_3898422994082169068pre" style="margin:0;padding:0;font-family:monospace">Programa de Pós-Graduação em Estatística<br> Universidade de Brasília<br> <a href="http://www.pgest.unb.br" rel="noreferrer" target="_blank">www.pgest.unb.br</a><br> Telefone: (61) 3107-3697</div>
</font></span></div>
</div>
</div><br></div>