Research map / Learning theory and algorithms / Sample complexity
Two is Better Than One: Regularized Shrinkage of Large Minimum Variance Portfolios
Taras Bodnar, Nestor Parolya, Erik Thorsen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use double shrinkage instead of BPS linear shrinkage (ec target) | GMV portfolio · S&P500, n=250, p=260 | out-of-sample std dev: 0.03912 -> 0.01104 | Improved |
| use double shrinkage instead of traditional GMV | GMV portfolio · S&P500, n=500, p=431 | out-of-sample std dev: 0.02591 -> 0.01049 | Improved |
| use double shrinkage instead of LW2020 nonlinear shrinkage | GMV portfolio · S&P500, n=250, p=260 | out-of-sample std dev: 0.01165 -> 0.01104 | Improved |
| use double shrinkage instead of LW2020 | GMV portfolio · S&P500, n=500, p=431 | out-of-sample std dev: 0.01099 -> 0.01049 | Improved |
| use double shrinkage instead of LW2020 for Sharpe ratio | GMV portfolio · S&P500, n=250, p=260 | Sharpe ratio: 0.047 -> 0.036 | Hurt |
| use equally weighted target instead of equal correlation target | Double shrinkage · S&P500, n=250, p=260 | turnover: 69.74 -> 3.36 | Improved |
| use equal correlation instead of equally weighted target | Double shrinkage · S&P500, n=250, p=260 | out-of-sample std dev: 0.01129 -> 0.01104 | Improved |
| use double shrinkage instead of BPS in index tracking | GMV portfolio · S&P500 index tracking, n=240 | turnover: 2555.37 -> 185.93 | Improved |
| use double shrinkage instead of BPS in index tracking | GMV portfolio · S&P500 index tracking, n=240 | out-of-sample std dev: 0.00834 -> 0.00913 | Hurt |
| use double shrinkage instead of BPS in index tracking | GMV portfolio · S&P500 index tracking, n=720 | Sharpe ratio: 0.028 -> 0.047 | Improved |
| increase window from n=250 to n=500 | Double shrinkage ec · S&P500, p=431 | out-of-sample std dev: 0.01088 -> 0.01049 | Improved |
| increase window from n=240 to n=720 | Double shrinkage · S&P500 index tracking | turnover: 185.93 -> 143.85 | Improved |
Closest papers by meaning
- Loss Control with Rank-one Covariance Estimate for Short-term Portfolio Optimization (JMLR 2020 · similarity 0.87)
- Latent Variable Estimation in Bayesian Black-Litterman Models (ICML 2025 · similarity 0.83)
- A Globally Optimal Portfolio for m-Sparse Sharpe Ratio Maximization (NeurIPS 2024 · similarity 0.82)
- Autonomous Sparse Mean-CVaR Portfolio Optimization (ICML 2024 · similarity 0.82)
- Automated Efficient Estimation using Monte Carlo Efficient Influence Functions (NeurIPS 2024 · similarity 0.81)
- On the Tightness and Computational Tractability of Higher-Dimensional Confidence Sequences (NeurIPS 2026 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
