Quantitative Models and Construction Methods 1. Model Name: Maximized Factor Exposure Portfolio (MFE) - Model Construction Idea: The MFE portfolio is designed to test the effectiveness of single factors under real-world constraints, such as industry exposure, style exposure, stock weight limits, and turnover rate. This approach ensures that the factors deemed "effective" can genuinely contribute to return prediction in the final portfolio[38][39]. - Model Construction Process: - The objective function is to maximize single-factor exposure, represented as $f^{T}w$, where $f$ is the factor value, and $w$ is the stock weight vector. - The optimization model includes the following constraints: 1. Style Exposure Constraint: Limits the portfolio's deviation from the benchmark in terms of style factors. $X$ is the factor exposure matrix, $w_b$ is the benchmark weight vector, and $s_l, s_h$ are the lower and upper bounds for style factor exposure[39]. 2. Industry Exposure Constraint: Limits the portfolio's deviation from the benchmark in terms of industry exposure. $H$ is the industry exposure matrix, and $h_l, h_h$ are the lower and upper bounds for industry exposure[39]. 3. Stock Weight Deviation Constraint: Limits individual stock weight deviations from the benchmark. $w_l, w_h$ are the lower and upper bounds for stock weight deviations[39]. 4. Constituent Stock Weight Constraint: Limits the weight of constituent stocks within the portfolio. $B_b$ is a binary vector indicating whether a stock is a benchmark constituent, and $b_l, b_h$ are the lower and upper bounds for constituent stock weights[39]. 5. No Short Selling Constraint: Ensures no short positions and limits individual stock weights to a maximum value $l$[39]. 6. Full Investment Constraint: Ensures the portfolio is fully invested, with the sum of weights equal to 1[40]. - The optimization model is expressed as: - The MFE portfolio is constructed monthly, and historical returns are backtested with a 0.3% transaction cost applied on both sides[42]. - Model Evaluation: The MFE portfolio effectively tests factor performance under realistic constraints, making it a robust tool for evaluating factor predictability in practical scenarios[38][39]. --- Factor Construction and Methods 1. Factor Name: EPTTM (Earnings to Price Trailing Twelve Months) - Factor Construction Idea: Measures the profitability of a company relative to its market value, using trailing twelve months' earnings[15]. - Factor Construction Process: - Formula: $EPTTM = \frac{\text{Net Income (TTM)}}{\text{Market Value}}$ - The numerator represents the trailing twelve months' net income, while the denominator is the company's total market value[15]. 2. Factor Name: BP (Book-to-Price Ratio) - Factor Construction Idea: Evaluates the valuation of a company by comparing its book value to its market value[15]. - Factor Construction Process: - Formula: $BP = \frac{\text{Book Value}}{\text{Market Value}}$ - The numerator is the company's book value, and the denominator is its total market value[15]. 3. Factor Name: Three-Month Volatility - Factor Construction Idea: Captures the stock's price fluctuation over the past three months, reflecting its risk level[15]. - Factor Construction Process: - Formula: $Volatility = \text{Average True Range (ATR)}$ over the past 60 trading days. - The ATR is calculated as the average of the daily high-low range over the specified period[15]. 4. Factor Name: One-Month Reversal - Factor Construction Idea: Measures the short-term reversal effect by analyzing the stock's return over the past month[15]. - Factor Construction Process: - Formula: $Reversal = \text{Return over the past 20 trading days}$ - Positive values indicate a reversal effect, while negative values suggest momentum continuation[15]. --- Factor Backtesting Results 1. EPTTM - HS300: Weekly return 1.35%, monthly return 4.28%, YTD return 5.95%, historical annualized return 4.60%[18]. - CSI500: Weekly return 1.54%, monthly return 3.55%, YTD return -3.61%, historical annualized return 4.78%[20]. - CSI1000: Weekly return 1.44%, monthly return 2.78%, YTD return 0.15%, historical annualized return 6.84%[22]. - CSIA500: Weekly return 1.72%, monthly return 3.92%, YTD return 2.62%, historical annualized return 3.71%[24]. - Public Fund Index: Weekly return 1.82%, monthly return 5.32%, YTD return 4.75%, historical annualized return 1.42%[26]. 2. BP - HS300: Weekly return 1.25%, monthly return 2.83%, YTD return -1.86%, historical annualized return 2.72%[18]. - CSI500: Weekly return 1.36%, monthly return 2.23%, YTD return 3.09%, historical annualized return 3.47%[20]. - CSI1000: Weekly return 0.99%, monthly return 1.56%, YTD return -0.45%, historical annualized return 3.07%[22]. - CSIA500: Weekly return 1.50%, monthly return 3.44%, YTD return -4.52%, historical annualized return 2.89%[24]. - Public Fund Index: Weekly return 1.45%, monthly return 3.20%, YTD return -8.75%, historical annualized return 0.74%[26]. 3. Three-Month Volatility - HS300: Weekly return 0.52%, monthly return 1.75%, YTD return -3.56%, historical annualized return 1.84%[18]. - CSI500: Weekly return 1.76%, monthly return 3.07%, YTD return -7.17%, historical annualized return 3.50%[20]. - CSI1000: Weekly return 1.40%, monthly return 2.54%, YTD return -8.22%, historical annualized return 4.33%[22]. - CSIA500: Weekly return 0.79%, monthly return 2.15%, YTD return -9.34%, historical annualized return 2.77%[24]. - Public Fund Index: Weekly return 0.97%, monthly return 2.04%, YTD return -15.34%, historical annualized return 1.54%[26]. 4. One-Month Reversal - HS300: Weekly return -0.93%, monthly return 0.98%, YTD return -0.57%, historical annualized return -0.33%[18]. - CSI500: Weekly return -1.83%, monthly return -0.84%, YTD return 2.56%, historical annualized return -0.84%[20]. - CSI1000: Weekly return -1.49%, monthly return -0.55%, YTD return -4.63%, historical annualized return -3.84%[22]. - CSIA500: Weekly return -1.28%, monthly return 0.51%, YTD return -1.07%, historical annualized return -2.34%[24]. - Public Fund Index: Weekly return -1.11%, monthly return 0.95%, YTD return 4.67%, historical annualized return -1.80%[26].
多因子选股周报:估值因子表现出色,沪深 300 指增组合年内超额18.92%-20251108
Guoxin Securities·2025-11-08 12:08