基于分位数回归的两步估计稀疏指数追踪
Two-Step Estimation Sparse Index Tracking Based on Quantile Regression
DOI: 10.12677/ORF.2023.134355, PDF,   
作者: 马 林：贵州大学数学与统计学院，贵州 贵阳
关键词: 指数追踪；稳健估计；分位数回归；两步估计；Index Tracking； Robust Estimation； Quantile Regression； Two-Step Estimation
摘要: 指数追踪这种被动管理方法凭借其风险小成本低的优势，受到大量投资者的追捧，其目标是最小化追踪误差。本文采用部分复制的策略对上证180指数进行指数追踪，以均方根误差作为衡量标准。利用表现较好的两步估计方法来追踪指数，第一步利用弹性网进行变量选择，第二步考虑模型的稳健性使用分位数回归来确定系数。实证分析结果显示两步估计方法优于单一方法，而基于分位数回归两步估计是其中表现最好的，因此可以使用此模型来进行指数追踪。
Abstract: Index tracking is a passive management method that is sought after by a large number of investors due to its low risk and low cost, and its goal is to minimize tracking errors. In this paper, the index tracking of the SSE 180 Index is carried out by a partially replicated strategy, and the root mean square error is used as the measurement standard. The index is tracked using a well-performing two-step estimation method, the first step using the elastic net for variable selection, and the second step considering the robustness of the model using quantile regression to determine the coefficients. Empirical analysis results show that the two-step estimation method is superior to the single-step method, and the two-step estimation based on quantile regression is the best, so this model can be used for exponential tracking.
文章引用：马林. 基于分位数回归的两步估计稀疏指数追踪[J]. 运筹与模糊学, 2023, 13(4): 3527-3535. https://doi.org/10.12677/ORF.2023.134355

参考文献

[1] Kim, S. and Kim, S. (2020) Index Tracking through Deep Latent Representation Learning. Quantitative Finance, 20, 639-652. [Google Scholar] [CrossRef] 
[2] Kwak, Y., Song, J. and Lee, H. (2021) Neural Network with Fixed Noise for Index-Tracking Portfolio Optimization. Expert Systems with Applications, 183, 115298. [Google Scholar] [CrossRef] 
[3] Bradrania, R., Pirayesh Neghab, D. and Shafizadeh, M. (2022) State-Dependent Stock Selection in Index Tracking: A Machine Learning Approach. Financial Markets and Portfolio Management, 36, 1-28. [Google Scholar] [CrossRef] 
[4] Cao, Y., Li, H. and Yang, Y. (2022) Combining Random Forest and Multicollinearity Modeling for Index Tracking. Communications in Statistics-Simulation and Computa-tion, 1-12. [Google Scholar] [CrossRef] 
[5] Wu, L., Yang, Y. and Liu, H. (2014) Nonnegative-Lasso and Application in Index Tracking. Computational Statistics & Data Analysis, 70, 116-126. [Google Scholar] [CrossRef] 
[6] Wu, L. and Yang, Y. (2014) Nonnegative Elastic Net and Application in Index Tracking. Applied Mathematics and Computation, 227, 541-552. [Google Scholar] [CrossRef] 
[7] Yang, Y. and Wu, L. (2016) Nonnegative Adaptive Lasso for Ultra-High Dimensional Regression Models and a Two- Stage Method Applied in Financial Modeling. Journal of Statistical Planning and Inference, 174, 52-67. [Google Scholar] [CrossRef] 
[8] Li, N., Yang, H. and Yang, J. (2021) Nonnegative Estimation and Variable Selection via Adaptive Elastic-Net for High- Dimensional Data. Communications in Statis-tics-Simulation and Computation, 50, 4263-4279. [Google Scholar] [CrossRef] 
[9] Li, N. and Yang, H. (2021) Nonnegative Estimation and Variable Selection under Minimax Concave Penalty for Sparse High-Dimensional Linear Regression Models. Sta-tistical Papers, 62, 661-680. [Google Scholar] [CrossRef] 
[10] Chen, Q., Hu, Q., Yang, H., et al. (2022) A Kind of New Time-Weighted Nonnegative Lasso Index-Tracking Model and Its Application. The North American Journal of Economics and Finance, 59, 101603. [Google Scholar] [CrossRef] 
[11] Li, N. (2020) Efficient Sparse Portfolios Based on Composite Quantile Regression for High-Dimensional Index Tracking. Journal of Statistical Computation and Simulation, 90, 1466-1478. [Google Scholar] [CrossRef] 
[12] Li, N., Niu, Y. and Sun, J. (2022) Robust Sparse Port-folios for Index Tracking Based on M-Estimation. Communications in Statistics-Simulation and Computation, 1-13. [Google Scholar] [CrossRef] 

为你推荐