Penalized Likelihood Inference with Survey Data

• Tác giả: Joann Jasiak, Purevdorj Tuvaandorj
• Ngôn ngữ: eng
• Ký hiệu phân loại: 001.433 Descriptive method
• Thông tin xuất bản: 2023
• Bộ sưu tập: Metadata
• ID: 196924

Tóm tắt

Comment: 56 pagesThis paper extends three Lasso inferential methods, Debiased Lasso, $C(\alpha)$ and Selective Inference to a survey environment. We establish the asymptotic validity of the inference procedures in generalized linear models with survey weights and/or heteroskedasticity. Moreover, we generalize the methods to inference on nonlinear parameter functions e.g. the average marginal effect in survey logit models. We illustrate the effectiveness of the approach in simulated data and Canadian Internet Use Survey 2020 data.