Sharpe-Ratio-Oriented Securities Portfolio Simulation Optimization: A Case Study Based on the A-Share Market

Abstract

As financial market volatility increases, individual investors are paying greater attention to risk-adjust returns. Modern investment theory emphasizes balancing risk and returns under uncertainty. In portfolio investment, the Sharpe ratio is a very important theoretical foundation. The Sharpe ratio describes risk-adjusted performance by “excess return per unit of volatility,” and because it is highly comparable, it is widely used in the evaluation of funds and investment portfolios. Taking the A-share market as the background, this paper constructs a securities portfolio simulation optimization framework with “maximizing the Sharpe ratio” as the core objective. Based on mean–variance theory, it transforms the fractional objective into a solvable constrained optimization problem, and incorporates realistic conditions such as long-only constraints, weight caps, and transaction costs. Using rolling-window out-of-sample back testing, it compares the differences between two risk estimation methods—sample covariance and shrinkage covariance—in terms of weight stability and risk-adjusted returns. Considering that the A-share trading system has features such as order size units and daily price limit constraints, this paper discusses tradability constraints in the back-testing assumptions. The simulation results show that the shrinkage covariance approach has advantages in reducing extreme weights and turnover, and its out-of-sample Sharpe ratio and maximum drawdown are also more robust. The contribution of this paper lies in integrating “Sharpe objective—estimation error—market institutional frictions” into a unified, reproducible research paradigm, providing a clear writing and empirical path for portfolio optimization research in the A-share environment.