东方因子周报：Value风格登顶，预期EPTTM因子表现出色-20250615

Quantitative Factors and Models Summary Quantitative Factors and Construction Process - Factor Name: Value Construction Idea: Captures the market's preference for value investing strategies Construction Process: Includes metrics like Book-to-Price (BP) and Earnings Yield (EP) Evaluation: Demonstrated significant positive returns in the recent week, indicating increased market recognition of value strategies [10][12][13] - Factor Name: SOE (State-Owned Enterprise) Construction Idea: Measures the proportion of state ownership in a company Construction Process: Calculated as the percentage of shares held by state-owned entities Evaluation: Showed slight improvement in returns, reflecting increased market attention to state-owned enterprises [10][13] - Factor Name: Growth Construction Idea: Tracks growth potential through metrics like ROE changes and sales growth Construction Process: Includes metrics such as Delta ROE (average ROE change over 3 years) and Sales Growth (3-year compound growth rate of revenue) Evaluation: Stable performance, though slightly weaker compared to the previous week [11][13] - Factor Name: Volatility Construction Idea: Measures the market's preference for high or low volatility stocks Construction Process: Includes metrics like standard deviation of returns over 243 days and maximum/minimum returns over the same period Evaluation: Negative returns, indicating reduced demand for high-volatility assets [11][13] - Factor Name: Trend Construction Idea: Captures momentum through exponential weighted moving averages (EWMA) Construction Process: Metrics include Trend_120 (EWMA with half-life of 20 days divided by EWMA with half-life of 120 days) Evaluation: Significant decline in returns, reflecting reduced market preference for trend-following strategies [11][13] - Factor Name: Certainty Construction Idea: Measures market confidence in predictable investments Construction Process: Includes metrics like institutional holding percentage and analyst coverage Evaluation: Weak performance, indicating low market confidence in certainty-based strategies [11][13] - Factor Name: Liquidity Construction Idea: Tracks market preference for liquid assets Construction Process: Includes metrics like average turnover rate over 243 days and liquidity beta Evaluation: Declined further, showing reduced demand for high-liquidity stocks [11][13] - Factor Name: Beta Construction Idea: Measures sensitivity to market movements Construction Process: Bayesian compressed market beta Evaluation: Significant negative returns, indicating reduced preference for high-beta stocks [11][13] Factor Backtesting Results - Value Factor: Weekly return of 0.38%, monthly return of -4.25%, annualized return of -24.01% over the past year [12] - SOE Factor: Weekly return of 0.16%, monthly return of 0.71%, annualized return of 19.07% over the past year [12] - Growth Factor: Weekly return of -0.19%, monthly return of 2.21%, annualized return of 15.05% over the past year [12] - Volatility Factor: Weekly return of -0.28%, monthly return of -1.38%, annualized return of 24.81% over the past year [12] - Trend Factor: Weekly return of -0.41%, monthly return of 3.86%, annualized return of 16.35% over the past year [12] - Certainty Factor: Weekly return of -0.66%, monthly return of -3.15%, annualized return of -20.57% over the past year [12] - Liquidity Factor: Weekly return of -1.06%, monthly return of -2.38%, annualized return of 30.40% over the past year [12] - Beta Factor: Weekly return of -1.30%, monthly return of -0.80%, annualized return of 29.44% over the past year [12] MFE Portfolio Construction - Model Name: Maximized Factor Exposure (MFE) Portfolio Construction Idea: Maximizes single-factor exposure while controlling for constraints like industry and style deviations Construction Process: - Objective function: Maximize factor exposure - Constraints: - Style and industry exposure limits - Stock weight deviation limits - Turnover rate constraints - Formula: $ \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} \ & 0\leq w\leq l \ & 1^{T}w=1 \ & \Sigma|w-w_{0}|\leq to_{h} \end{array} $ - $f$: Factor values - $w$: Stock weight vector - $X$: Style exposure matrix - $H$: Industry exposure matrix - $B_b$: Benchmark component weights - $to_h$: Turnover rate upper limit Evaluation: Effective in isolating factor performance under controlled conditions [59][60][63] MFE Portfolio Backtesting Results - HS300 Index: Weekly excess return range: -0.41% to 0.72%, median: 0.21% [52] - CSI500 Index: Weekly excess return range: -0.55% to 1.25%, median: 0.28% [56] - CSI1000 Index: Weekly excess return range: -0.64% to 1.05%, median: 0.22% [58]