金工股票策略环境监控周报：本周宽基指数普涨但情绪降温近期可重点考虑投资组合的抗风险能力-20250915

Quantitative Models and Construction Methods 1. Model Name: Barra Style Factors - Model Construction Idea: The model aims to capture the performance of various style factors in the equity market, such as momentum, size, and residual volatility[12][27] - Model Construction Process: The model calculates the returns of different style factors over a specified period. For example, the momentum factor return is calculated as: $\text{Momentum Factor Return} = \frac{\sum (\text{Stock Returns} \times \text{Momentum Scores})}{\sum \text{Momentum Scores}}$ where the momentum scores are derived from the past performance of stocks[12][27] - Model Evaluation: The model effectively captures the performance of different style factors, providing insights into market trends and investor behavior[12][27] 2. Model Name: Excess Return Monitoring Model - Model Construction Idea: This model monitors the relative performance of small and mid-cap indices against a large-cap benchmark to capture market style rotation signals[86] - Model Construction Process: The model calculates the rolling 20-day excess returns of indices such as CSI 2000, CSI 1000, and CSI 500 relative to the CSI 300. For example: $\text{Excess Return} = \text{CSI 1000 Return} - \text{CSI 300 Return}$ The model then tracks the percentile rank of these excess returns over a three-year period to identify significant deviations[86] - Model Evaluation: The model provides a systematic approach to detect market style rotations, aiding in strategic asset allocation decisions[86] Model Backtest Results - Barra Style Factors: - Momentum Factor: Weekly return 0.61%, monthly return 0.96%, annualized Sharpe ratio 2.20[12][27] - Size Factor: Weekly return 0.56%, monthly return 1.71%, annualized Sharpe ratio -1.78[12][27] - Residual Volatility Factor: Weekly return -0.48%, monthly return -0.77%, annualized Sharpe ratio -1.65[12][27] - Excess Return Monitoring Model: - CSI 1000 vs. CSI 300: 20-day rolling return -3.32%, 3-year percentile 17.0%[86] - CSI 2000 vs. CSI 300: 20-day rolling return -4.78%, 3-year percentile 15.1%[86] - CSI 500 vs. CSI 300: 20-day rolling return 1.21%, 3-year percentile 66.7%[86] Quantitative Factors and Construction Methods 1. Factor Name: Momentum - Factor Construction Idea: The momentum factor captures the tendency of stocks that have performed well in the past to continue performing well in the future[12][27] - Factor Construction Process: The momentum score for each stock is calculated based on its past returns over a specified period, typically 12 months. The factor return is then computed as: $\text{Momentum Factor Return} = \frac{\sum (\text{Stock Returns} \times \text{Momentum Scores})}{\sum \text{Momentum Scores}}$ where the momentum scores are derived from the past performance of stocks[12][27] - Factor Evaluation: The momentum factor has shown consistent positive returns, indicating its effectiveness in capturing market trends[12][27] 2. Factor Name: Size - Factor Construction Idea: The size factor captures the performance difference between small-cap and large-cap stocks[12][27] - Factor Construction Process: The size score for each stock is calculated based on its market capitalization. The factor return is then computed as: $\text{Size Factor Return} = \frac{\sum (\text{Stock Returns} \times \text{Size Scores})}{\sum \text{Size Scores}}$ where the size scores are derived from the market capitalization of stocks[12][27] - Factor Evaluation: The size factor has shown mixed performance, reflecting the varying investor preferences for small-cap versus large-cap stocks over time[12][27] Factor Backtest Results - Momentum Factor: Weekly return 0.61%, monthly return 0.96%, annualized Sharpe ratio 2.20[12][27] - Size Factor: Weekly return 0.56%, monthly return 1.71%, annualized Sharpe ratio -1.78[12][27] - Residual Volatility Factor: Weekly return -0.48%, monthly return -0.77%, annualized Sharpe ratio -1.65[12][27]