Loading...
|
Please use this identifier to cite or link to this item:
https://nccur.lib.nccu.edu.tw/handle/140.119/152780
|
| Title: | 基於Copula Entropy的變數選取方法與節點選取方法
Variable Selection Method and Knot Selection Method Based on Copula Entropy |
| Authors: | 張軼棠
Chang, I-Tang |
| Contributors: | 黃子銘
Huang, Tzee-Ming
張軼棠
Chang, I-Tang |
| Keywords: | Lasso變數選取
Stepwise變數選取
Copula entropy
B-spline函數
Lasso Variable Selection
Stepwise Variable Selection
Copula Entropy
B-spline |
| Date: | 2024 |
| Issue Date: | 2024-08-05 14:00:15 (UTC+8) |
| Abstract: | 本研究透過copula entropy的應用，優化stepwise變數選取方法，並將其應用於B-spline函數中來近似實際函數，以挑選出必要的節點，同時又能減少與實際函數間的誤差。此方法利用copula entropy的獨特特性，以及RCV(refitted cross-validation)變異數估計方法，改善了stepwise變數選取的方法，此外，我們將改進方法與其他方法進行比較，以驗證其在實際應用中的效能表現。實驗結果顯示，此方法在回歸函數上誤差和準確率方面優於其他常見的變數選取方法，在近似spline函數上於部分情況中也表現出較佳的節點挑選效果，進而在B-spline函數的應用中實現更有效率的節點選擇。
In this study, we optimize the stepwise variable selection method through the application of copula entropy and apply it to the B-spline function to approximate the actual function in order to pick out the necessary knots and at the same time reduce the error with the actual function. This method improves the stepwise variable selection method by utilizing the unique characteristics of copula entropy and the RCV (refitted cross-validation) estimation method. In addition, we compare the improved method with other methods to verify its performance in practical applications. The experimental results show that this method outperforms other common variable selection methods in terms of error and accuracy on the regression function, and also shows better knot selection on the approximate spline function in some cases, which leads to more efficient knot selection in the application of the B-spline function. |
| Reference: | Ty Adams. Forward selection via distance correlation. 2019.
Jianqing Fan, Shaojun Guo, and Ning Hao. Variance Estimation Using Refitted Cross-Validation in Ultrahigh Dimensional Regression. Journal of the Royal Statistical Society
Series B: Statistical Methodology, 74(1):37–65, 10 2011.
Arthur E. Hoerl and Robert W. Kennard. Ridge regression: Biased estimation for nonorthogonal problems. Technometrics, 42(1):80–86, 2000.
Alexander Kraskov, Harald Stögbauer, and Peter Grassberger. Estimating mutual information. Physical Review E, 69(6), June 2004.
Jian Ma. Variable selection with copula entropy. ArXiv, abs/1910.12389, 2019.
Jian Ma. copent: Estimating copula entropy and transfer entropy in r, 2021.
E Sunandi, K A Notodoputro, and B Sartono. A study on group lasso for grouped variable selection in regression model. IOP Conference Series: Materials Science and Engineering, 1115(1):012089, 2021.
Robert Tibshirani. Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society. Series B (Methodological), 58(1):267–288, 1996.
Ming Yuan and Yi Lin. Model selection and estimation in regression with grouped variables. Journal of the Royal Statistical Society Series B, 68:49–67, 2006.
Dennis D Boos Yujun Wu and Leonard A Stefanski. Controlling variable selection by the addition of pseudovariables. Journal of the American Statistical Association,
102(477):235–243, 2007.
Hui Zou. The adaptive lasso and its oracle properties. Journal of the American Statistical Association, 101(476):1418–1429, 2006.
Hui Zou and Trevor Hastie. Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society. Series B (Statistical Methodology), 67(2):301–320, 2005. |
| Description: | 碩士
國立政治大學
統計學系
111354025 |
| Source URI: | http://thesis.lib.nccu.edu.tw/record/#G0111354025 |
| Data Type: | thesis |
| Appears in Collections: | [統計學系] 學位論文
|
Files in This Item:
| File |
Description |
Size | Format | |
|---|
| 402501.pdf | | 934Kb | Adobe PDF | 0 | View/Open |
|
All items in 政大典藏 are protected by copyright, with all rights reserved.
|
