金工定期报告：从微观出发的风格轮动月度跟踪

• The style rotation model is constructed based on micro-level stock selection factors, including valuation, market capitalization, volatility, and momentum. It utilizes 80 base factors to create 640 micro features, replacing absolute proportion division of style factors with common indices as style stock pools. A random forest model is employed for rolling training to avoid overfitting, enabling feature selection and style recommendation. The framework integrates style timing, scoring, and actual investment strategies[8]. - The performance of the style rotation model during the backtesting period (2014/01/01-2024/11/30) shows an annualized return of 22.97%, annualized volatility of 24.55%, IR of 0.94, monthly win rate of 60.31%, and maximum drawdown of 28.33%. When hedged against the market benchmark, the annualized return is 14.07%, annualized volatility is 11.62%, IR is 1.21, monthly win rate is 67.94%, and maximum drawdown is 8.98%[9][10]. - In 2024, the strategy achieved a return of 28.59%, outperforming the market benchmark by 10.53%[15]. For November 2024, the model's return was 1.52%, with an excess return of 0.84% relative to the benchmark[16]. The latest style timing directions for December 2024 are value, small-cap, momentum, and low volatility[17]. The current holding index is the Emerging Growth Index[18].