| dc.contributor | 統計系 | |
| dc.creator (作者) | 陳立榜 | |
| dc.creator (作者) | Chen, Li-Pang;Wu, Jou-Chin;Tsao, Hui-Shan | |
| dc.date (日期) | 2026-04 | |
| dc.date.accessioned | 5-May-2026 13:59:16 (UTC+8) | - |
| dc.date.available | 5-May-2026 13:59:16 (UTC+8) | - |
| dc.date.issued (上傳時間) | 5-May-2026 13:59:16 (UTC+8) | - |
| dc.identifier.uri (URI) | https://ah.lib.nccu.edu.tw/item?item_id=182292 | - |
| dc.description.abstract (摘要) | Discriminant analysis has been a commonly used strategy to handle classification with binary or multi-label responses. Under multivariate normal distributions of covariates, a linear or quadratic discriminant function can be derived, which is used as a boundary to classify subjects. In the discriminant function, the estimation of the precision matrix, which is defined as the inverse of the covariance matrix, is a crucial issue because of the sparsity, reflecting that few pairs of variables are informative, which may implicitly affect the accuracy of the classification. While a large body of estimation methods is available to estimate the precision matrix, most methods not only fail to handle ultrahigh-dimensionality in the sense that the dimension of variables is extremely larger than the sample size, but also require longer computational time. In addition, in the presence of nonlinear dependence among variables, existing methods may falsely miss the detection of dependence. To tackle those challenges, we extend the model-free feature screening method to reduce the dimension of variables and detect possibly nonlinear pairwise dependence structure among variables. After that, we adopt several graphical estimation methods to estimate the homogeneous or heterogeneous precision matrices and then implement the estimator to the discriminant function to improve the classification. Numerical studies are conducted to assess the performance of prediction for the proposed method and verify the necessity of taking network structures of variables into account. | |
| dc.format.extent | 108 bytes | - |
| dc.format.mimetype | text/html | - |
| dc.relation (關聯) | Big Data Analytics in Biostatistics and Bioinformatics. ICSA Book Series in Statistics, Springer, pp.215-244 | |
| dc.subject (關鍵詞) | Bioinformatics; Classification; Feature screening; Graphical models; Precision matrix | |
| dc.title (題名) | Ultrahigh-dimensional discriminant analysis and its application to gene expression data | |
| dc.type (資料類型) | book/chapter | |
| dc.identifier.doi (DOI) | 10.1007/978-3-032-06649-7_10 | |
| dc.doi.uri (DOI) | https://doi.org/10.1007/978-3-032-06649-7_10 | |
