American Statistical Association
Treatment decisions and prognosis frequently stem from statistical models that are developed to assess the chance of a patient response. Many of these models introduce a new biomarker with the implication that the additional marker represents improved patient risk classification. There exist many measures of model based risk classification, one of the oldest being a measure of discrimination termed the area under the curve (AUC). In this work, we develop an inferential procedure to determine whether the introduction of a new set of biomarkers provides a clinically relevant improvement in patient classification, as determined by the change in the area under the curve statistic. The asymptotic distribution for the change in AUC is derived, and as a result, an asymptotic confidence interval for this change parameter is computed.
|Date:||Wednesday, December 3, 2014|
|Time:||4:00 - 5:00 P.M.|
Memorial Sloan-Kettering Cancer Center
Department of Epidemiology and Biostatistics
307 East 63rd Street
(between First and Second Avenues)
3rd Floor Conference Room
New York, New York
Note: To gain access to the building, please follow the directions by the telephone in the foyer.