Graduate Student Seminar

Speaker: Melody Denhere

Title: Robust Estimation of Functional Logistic Regression Using Principal Component Estimation

Abstract: We propose a robust estimation technique for the parameter function of the functional logistic regression model. The proposed method is based on a robust principal component approach that allows for more efficient estimation of the parameter function in the presence of outlying curves. We demonstrate the performance of the proposed method in a simulation study that explores the proposed estimator's robustness properties. Several real data examples on near infrared spectroscopy data and monthly weather data are also used to illustrate the performance of the proposed method.

Note:  Melody is a Ph.D. student in our department; she is working with Dr. Billor.