Pei Geng: An integrated least square framework for Logistic regression with measurement error under case-control studies

Speaker: Pei Geng

Institution: University of New Hampshire Department of Mathematics and Statistics

Title: An integrated least square framework for Logistic regression with measurement error under case-control studies

Abstract: In Logistic regression under the case–control framework, the logarithmic ratio of the covariate densities between the case and control groups is a linear function of the regression parameters. Based on this fact, this talk introduces an integrated least-square-type framework for parameter estimation in Logistic regression based on the estimated covariate densities. If the covariate is precisely measured, the covariate densities can be effectively estimated by kernel density estimation. When the covariate is measured with errors, we develop two bias correction techniques to obtain consistent estimators: deconvolution kernel estimation and estimation using validation data. We conducted simulation study to evaluate the performance of the proposed method and further applied our method to the BRFSS and NHANES data to investigate the relationship between diabetes and BMI. 

Bio: Dr. Pei Geng is an assistant professor of Statistics at Department of Mathematics and Statistics, University of New Hampshire. She received her Ph.D. in Statistics from Michigan State University. Her research focuses on measurement error models, machine learning methods in genetic data analysis and survey sampling procedures.