多因子选股周报：反转因子表现出色，沪深300增强组合年内超额17.58%-20251018

• The report tracks the performance of Guosen Financial Engineering's index enhancement portfolios, which are constructed based on multi-factor stock selection models targeting benchmarks such as CSI 300, CSI 500, CSI 1000, and CSI A500 indices. The goal is to consistently outperform the respective benchmarks [11][12][14] - The construction process of the index enhancement portfolios includes three main components: return prediction, risk control, and portfolio optimization. The optimization model maximizes single-factor exposure while controlling for constraints such as industry exposure, style exposure, stock weight deviation, turnover rate, and component stock weight ratio [12][41][42] - The Maximized Factor Exposure (MFE) portfolio is used to test the effectiveness of individual factors under real-world constraints. The optimization model is expressed as follows: $\begin{array}{ll}max&f^{T}\ w\ s.t.&s_{l}\leq X(w-w_{b})\leq s_{h}\ &h_{l}\leq H(w-w_{b})\leq h_{h}\ &w_{l}\leq w-w_{b}\leq w_{h}\ &b_{l}\leq B_{b}w\leq b_{h}\ &\mathbf{0}\leq w\leq l\ &\mathbf{1}^{T}\ w=1\end{array}$ where $f$ represents factor values, $w$ is the stock weight vector, and constraints include style exposure ($X$), industry exposure ($H$), stock weight deviation ($w_b$), and component stock weight ratio ($B_b$) [41][42][43] - The report monitors the performance of common stock selection factors across different sample spaces, including CSI 300, CSI 500, CSI 1000, CSI A500, and public fund heavy-holding indices. Factors are tested using MFE portfolios to evaluate their excess return relative to benchmarks [11][15][18] - The factor library includes over 30 factors categorized into valuation, reversal, growth, profitability, liquidity, volatility, corporate governance, and analyst-related factors. Examples include BP (Book-to-Price), EPTTM (Earnings-to-Price TTM), one-month reversal, three-month reversal, one-year momentum, and others [16][17] - The report highlights the performance of specific factors in different sample spaces: - CSI 300: One-month reversal, three-month reversal, and EPTTM one-year percentile performed well recently, while three-month institutional coverage and standardized unexpected earnings performed poorly [1][18] - CSI 500: Three-month volatility, three-month reversal, and EPTTM one-year percentile performed well recently, while one-year momentum and standardized unexpected revenue performed poorly [1][20] - CSI 1000: One-month volatility, one-month turnover, and three-month reversal performed well recently, while executive compensation and three-month earnings revisions performed poorly [1][22] - CSI A500: One-month reversal, EPTTM one-year percentile, and one-month volatility performed well recently, while three-month institutional coverage and one-year momentum performed poorly [1][24] - Public fund heavy-holding index: Dividend yield, three-month reversal, and EPTTM performed well recently, while standardized unexpected revenue and three-month earnings revisions performed poorly [1][26][27] - The report tracks the excess returns of public fund index enhancement products, including CSI 300, CSI 500, CSI 1000, and CSI A500. For CSI 300 products, the highest weekly excess return was 0.92%, while the lowest was -3.08%, with a median of 0.01% [3][32][31] - For CSI 500 products, the highest weekly excess return was 3.20%, while the lowest was -0.48%, with a median of 0.49% [3][35][34] - For CSI 1000 products, the highest weekly excess return was 1.58%, while the lowest was -0.82%, with a median of 0.37% [3][37][36] - For CSI A500 products, the highest weekly excess return was 1.20%, while the lowest was -0.84%, with a median of 0.23% [3][40][39]