High-Dimensional portfolio selection with HDShOP package
本文介绍R包HDShOP，用于构建高维收缩投资组合，包括均值向量、协方差矩阵和精度矩阵的收缩估计，以及最优组合权重的直接收缩估计，并附有基于标普500股票数据的实证示例。
This paper discusses the practical aspects of working with high-dimensional shrinkage portfolios. It presents the R package HDShOP which provides a comprehensive framework for such work. In particular, we cover the construction of portfolios using shrinkage-based estimators for the mean vector, covariance matrix, and precision matrix of asset returns, as well as the shrinkage estimators derived directly for the weights of optimal portfolios. Moreover, shrinkage-based tests on the mean-variance efficiency of a given portfolio are discussed. Aspects related to programming, such as classes and methods used in the construction of optimal portfolios, are described. The description of the software is preceded by underlying theory and it is accompanied by several empirical illustrations based on the data consisting of returns on stocks from the S&P 500 index.