Frequentist model averaging in regression models
Speaker :
Time :
Location :
Title: Frequentistmodel averaging in regression models
Speaker: Dr. Shaobo Jin(金少博), Uppsala University
Time: 3:00pm, 2018-11-21
Location: The 4th Floor, Sir RRShaw Building, Yuquan Campus, Zhejiang University
Abstract: In many applications ofregression models, randomness due to model selection is commonly ignored inpost-model selection inference, which leads to too optimistic confidenceintervals. When the main focus is prediction, model selection uncertainty isalso often overlooked. In order to account for the model selection uncertainty,least-squares or likelihood-based frequentist model averaging has been recentlyproposed. Instead of choosing the optimal candidate model, a weighted averageof candidate models is preferred. In this talk, frequentist model averaging inregression models will be briefly reviewed. In the linear regression context,we can show that the confidence interval from model averaging is asymptotically equivalent to the confidence intervalfrom the full model. The asymptotic-based finite-sample confidence intervalsare equivalent to that from the full model if the parameter of interest is alinear function of the regression coefficients. We can also show that thisequivalence holds for prediction intervals. We can also apply the principle offrequentist model averaging to covariance analysis-based latent regressionmodels. Some current work will be briefly discussed.
All are welcome!
Contact Person:Zhonggen Su, suzhonggen@zju.edu.cn