<html><head><style type='text/css'>p { margin: 0; }</style></head><body><div style='font-family: times new roman,new york,times,serif; font-size: 14pt; color: #000000'>Caros redistas,<br><span class="Object" id="OBJ_PREFIX_DWT2367_com_zimbra_date"><span class="Object" id="OBJ_PREFIX_DWT2368_com_zimbra_date"></span></span><br><div>Na próxima sexta-feira daremos continuidade ao Ciclo de Seminários do Programa Interinstitucional de Pós-Graduação em Estatística (PIPGES) da UFSCar/USP.</div><br>05 de maio de 2017, sexta-feira, às 14h. Sala 43 do DEs-UFSCar.<br><br>PALESTRANTE: Manoel Ferreira dos Santos Neto<br><br>TÍTULO: Reparameterized Birnbaum-Saunders regression models with varying <br>precision<br><br>RESUMO:<br>We propose a methodology based on a reparameterized Birnbaum-Saunders <br>regression model with varying precision, which generalizes the existing <br>works in the literature on the topic. This methodology includes the <br>estimation of model parameters, hypothesis tests for the precision <br>parameter, a residual analysis and influence diagnostic tools. <br>Simulation studies are conducted to evaluate its performance. We apply <br>it to two real-world case-studies to show its potential with the R <br>software.<br><br>Saudações,<br>Ricardo<span></span><br><br></div></body></html>