2013 년 제 14 회 통계세미나 개최 안내
통계연구소에서는 다음과 같이 통계 세미나를 개최하오니 많은 참여 바랍니다 .
일시 : 2013 년 11 월 4 일 ( 월 ) 오후 5 시
장소 : 고려대학교 정경관 503 호
연사 : 한승봉 박사 ( 아산병원 )
Robust Semiparametric Variable Selection Using Predictive Accuracy for Interval-censored Data
<Abstract>
In establishing reliable prediction rules for future patients' prognosis, we need to consider which variables to include in the final prediction model. In many cohort studies, time to the occurrence of particular events such as disease recurrence is recorded in an interval-censored format. Clinicians are extremely interested in covariate combinations which can be used for the prediction model. To address this type of question, we proposed a covariate selection strategy to leverage the predictive accuracy based on a graphical measure and negative predictive functions. Simulation studies and an example from the breast cancer data further illustrate the practical advantages of this approach. The proposed semiparametric estimation of the predictive accuracy will provide simply and flexible approach for variable selection. The implementation effort is minimal, therefore this method has an increased potential for immediate use in biomedical data analyses. Considering the lack of public software availability for interval-censored data methodologies, this approach has an immediate practical advantage.
고려대학교 통계연구소