| dc.contributor | 統計系 | |
| dc.creator (作者) | 陳立榜 | |
| dc.creator (作者) | Chen, Li-Pang;Hsu, Wei-Hsin | |
| dc.date (日期) | 2023-09 | |
| dc.date.accessioned | 13-Dec-2023 13:55:00 (UTC+8) | - |
| dc.date.available | 13-Dec-2023 13:55:00 (UTC+8) | - |
| dc.date.issued (上傳時間) | 13-Dec-2023 13:55:00 (UTC+8) | - |
| dc.identifier.uri (URI) | https://nccur.lib.nccu.edu.tw/handle/140.119/148694 | - |
| dc.format.extent | 106 bytes | - |
| dc.format.mimetype | text/html | - |
| dc.relation (關聯) | Japanese Journal of Statistics and Data Science (Invited submission for the special issue: Recent Advances in Biostatistics) | |
| dc.subject (關鍵詞) | In this paper, we study causal inference with complex and noisy data accommodated. A new structure is called CHEMIST, which refers to Causal inference with High-dimensional Error-prone covariates and MISclassified Treatments. To suitably tackle those challenges when estimating the average treatment effect (ATE), we develop the FATE method, which reflects Feature screening, Adaptive lasso, Treatment adjustment, and Error elimination in covariates, to handle variable selection and measurement error correction. Under informative and error-eliminated data, we can estimate the ATE. To make our strategy available for public use, we develop a new R package CHEMIST, which provides functions for users to estimate the ATE. With the flexibility of arguments, one can examine different scenarios based on our package. In this paper, we introduce the FATE method and the implementation in the R package CHEMIST. Moreover, we demonstrate applications in two real data sets. | |
| dc.title (題名) | CHEMIST: an R package for causal inference with high-dimensional error-prone covariates and misclassified treatments | |
| dc.type (資料類型) | article | |
| dc.identifier.doi (DOI) | 10.1007/s42081-023-00217-y | |
| dc.doi.uri (DOI) | https://doi.org/10.1007/s42081-023-00217-y | |
