In high-dimensional data analysis, many penalized methods are introduced for simultaneous variable selection and parameter estimation when the model is sparse. However, a model may have sparse signals as well as predictors with weak signals. In this scenario, variable selection methods may not distinguish predictors with weak signals from those with sparse signals. For this reason, we propose post-shrinkage strategies to improve the prediction performance of a selected submodel, and the relative performance of the proposed strategy is appraised by theoretical simulation studies and real data analysis, respectively.
Guest speaker: Syed Ejaz Ahmed, PhD, Professor of Math & Statistics, Brock University
Location: Mission Hall #2700 or via ZOOM (Register Here if attending via ZOOM)