Sparsity and stability for minimum-variance portfolios

Abstract The popularity of modern portfolio theory has decreased among practitioners because its unfavorable out-of-sample performance. Estimation risk tends to affect the optimal weight calculation noticeably, especially when a large number assets are considered. To overcome these issues, many methods have been proposed in recent years, but only few address practically relevant questions related allocation. This study therefore uses different covariance estimation techniques, combines them with sparse model approaches, and includes turnover constraint that induces stability. We use two datasets S&P 500 create realistic data foundation for our empirical study. discover it is possible maintain low-risk profile efficient while automatically selecting subset further inducing low turnover. Moreover, we find simply using LASSO insufficient lower model’s tuning parameter can change over time.