<div dir="ltr"><br><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">---------- Forwarded message ---------<br>From: <strong class="gmail_sendername" dir="auto">Rodrigo Pescim</strong> <span dir="auto"><<a href="mailto:rrpescim@uel.br">rrpescim@uel.br</a>></span><br>Date: Tue, Mar 23, 2021 at 4:04 PM<br>Subject: Palestra - PGMAC/UEL - Prof. Rafael Izbicki<br>To: clarice demetrio <<a href="mailto:clarice.demetrio@usp.br">clarice.demetrio@usp.br</a>><br></div><br><br><div dir="ltr"><div style="margin-bottom:8px;font-family:"Segoe UI Historic","Segoe UI",Helvetica,Arial,sans-serif;color:rgb(5,5,5);font-size:15px;white-space:pre-wrap"><div style="font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">Olá Profa Clarice, tudo bem? </span></span></div><div style="font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit"><br></span></span></div><div style="font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">A profa poderia, por gentileza, fazer a divulgação de um webinário do programa de pós em matemática aplicada e computacional da UEL ? </span></span></div><div style="font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit"><br></span></span></div><div style="font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">Obrigado. </span></span></div><div style="font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit"><br></span></span></div><div style="font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">Abraços,</span></span></div><div style="font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit"><br></span></span></div><div style="font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">Rodrigo</span></span></div><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit"><br></span></span></div><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">#########</span></span></div><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit"><br></span></span></div><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">O ciclo de palestras do Programa de Verão 2021 - Linha de pesquisa em Modelagem Estatística - PGMAC/UEL convida a todos para o Webinário que será ministrado pelo Prof. Rafael Izbicki da UFSCar na próxima sexta-feira (26/03/2021) às 14h00.</span></span></div></div><div style="margin-bottom:8px;font-family:"Segoe UI Historic","Segoe UI",Helvetica,Arial,sans-serif;color:rgb(5,5,5);font-size:15px;white-space:pre-wrap"><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><br></span></div></div><div style="margin-bottom:8px;font-family:"Segoe UI Historic","Segoe UI",Helvetica,Arial,sans-serif;color:rgb(5,5,5);font-size:15px;white-space:pre-wrap"><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">TÍTULO: Quantification under prior probability shift: the ratio estimator and its extensions</span></span></div></div><div style="margin-bottom:8px;font-family:"Segoe UI Historic","Segoe UI",Helvetica,Arial,sans-serif;color:rgb(5,5,5);font-size:15px;white-space:pre-wrap"><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><br></span></div></div><div style="margin-bottom:8px;font-family:"Segoe UI Historic","Segoe UI",Helvetica,Arial,sans-serif;color:rgb(5,5,5);font-size:15px;white-space:pre-wrap"><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">RESUMO: The quantification problem consists of determining the prevalence of a given label in a target population. However, one often has access to the labels in a sample from the training population but not in the target population. A common assumption in this situation is that of prior probability shift, that is, once the labels are known, the distribution of the features is the same in the training and target populations. In this paper, we derive a new lower bound for the risk of the quantification problem under the prior shift assumption. Complementing this lower bound, we present a new approximately minimax class of estimators, ratio estimators, which generalize several previous proposals in the literature. Using a weaker version of the prior shift assumption, which can be tested, we show that  ratio estimators can be used to build confidence intervals for the quantification problem. We also extend the ratio estimator so that it can: (i) incorporate labels from the target population, when they are available and (ii) estimate how the prevalence of positive labels varies according to a function of certain covariates.</span></span></div></div><div style="margin-bottom:8px;font-family:"Segoe UI Historic","Segoe UI",Helvetica,Arial,sans-serif;color:rgb(5,5,5);font-size:15px;white-space:pre-wrap"><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">DOCENTE RESPONSÁVEL: Prof. Dr. Rafael Izbicki</span></span></div></div><div style="margin-bottom:8px;font-family:"Segoe UI Historic","Segoe UI",Helvetica,Arial,sans-serif;color:rgb(5,5,5);font-size:15px;white-space:pre-wrap"><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">DATA: 26/03/2021</span></span></div></div><div style="margin-bottom:8px;font-family:"Segoe UI Historic","Segoe UI",Helvetica,Arial,sans-serif;color:rgb(5,5,5);font-size:15px;white-space:pre-wrap"><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">LOCAL: </span></span><span style="font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit"><a href="http://meet.google.com/smd-iinh-dej" target="_blank">http://meet.google.com/smd-iinh-dej</a></span></span></span></div></div><div style="margin-bottom:8px;font-family:"Segoe UI Historic","Segoe UI",Helvetica,Arial,sans-serif;color:rgb(5,5,5);font-size:15px;white-space:pre-wrap"><div style="direction:ltr;font-family:inherit"><span style="font-family:inherit"><span style="font-family:inherit">HORÁRIO: 14:00Hrs</span></span></div></div><div><br></div>-- <br><div dir="ltr" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div dir="ltr"><span style="font-size:12.8px"><font color="#000000">Prof. Dr. Rodrigo Rossetto Pescim</font></span></div><div dir="ltr"><div style="font-size:12.8px"><font color="#000000"><span style="font-size:12.8px">Departamento de Estatística </span><br></font></div><div style="font-size:12.8px"><span style="color:rgb(0,0,0);font-size:12.8px">Universidade Estadual de Londrina - UEL</span><br></div><div style="font-size:12.8px"><span style="color:rgb(0,0,0);font-size:12.8px">Fone: (43) 3371-4932</span><span style="color:rgb(0,0,0);font-size:12.8px"><br></span></div><div style="font-size:12.8px"><span style="font-size:12.8px"><font color="#000000">Londrina - PR</font></span></div><div><span style="font-size:12.8px"><font color="#000000"><br></font></span></div></div></div></div></div></div></div>
</div></div>