国君金工|11月小市值风格强势,12月有望切换成大市值风格占优

Quantitative Models and Construction Methods 1. Model Name: Multi-Factor Model for Covariance Matrix Estimation - Model Construction Idea: The model aims to estimate the stock covariance matrix, which is critical for portfolio risk forecasting, by decomposing it into a factor covariance matrix and an idiosyncratic risk matrix[1] - Model Construction Process: - The covariance matrix of stocks is expressed as a combination of two components: $ \Sigma = F \cdot B \cdot F^T + D $ where: - $ \Sigma $ represents the stock covariance matrix - $ F $ is the factor covariance matrix - $ B $ is the factor loading matrix - $ D $ is the idiosyncratic risk matrix - The factor covariance matrix ($ F $) is derived from the returns of the factors, while the idiosyncratic risk matrix ($ D $) is estimated from the residuals of the multi-factor model regression[1] --- Quantitative Factors and Construction Methods 1. Factor Name: Volatility Factor - Factor Construction Idea: Measures the impact of stock price volatility on returns, with higher volatility stocks potentially offering higher returns in certain market conditions[1][2] - Factor Construction Process: - The factor is calculated based on the historical standard deviation of stock returns over a specific time window - Formula: $ \text{Volatility} = \sqrt{\frac{1}{N-1} \sum_{i=1}^{N} (R_i - \bar{R})^2} $ where: - $ R_i $ is the return of the stock on day $ i $ - $ \bar{R} $ is the average return over the period - $ N $ is the number of days in the period[1][2] 2. Factor Name: Size Factor (Market Capitalization) - Factor Construction Idea: Captures the relationship between a stock's market capitalization and its returns, with smaller-cap stocks often outperforming larger-cap stocks in certain periods[1][2] - Factor Construction Process: - The factor is calculated as the natural logarithm of the market capitalization of a stock - Formula: $ \text{Size} = \ln(\text{Market Cap}) $ where: - $ \text{Market Cap} $ is the stock's market capitalization[1][2] 3. Factor Name: Liquidity Factor - Factor Construction Idea: Reflects the impact of stock liquidity on returns, with less liquid stocks potentially offering higher returns due to liquidity risk premium[1][2] - Factor Construction Process: - The factor is calculated using the average daily trading volume over a specific period - Formula: $ \text{Liquidity} = \frac{\text{Average Daily Volume}}{\text{Market Cap}} $ where: - $ \text{Average Daily Volume} $ is the average number of shares traded daily - $ \text{Market Cap} $ is the stock's market capitalization[1][2] 4. Factor Name: Dividend Yield Factor - Factor Construction Idea: Measures the relationship between a stock's dividend yield and its returns, with higher dividend yield stocks often preferred in certain market conditions[2] - Factor Construction Process: - The factor is calculated as the annual dividend per share divided by the stock price - Formula: $ \text{Dividend Yield} = \frac{\text{Annual Dividend Per Share}}{\text{Stock Price}} $[2] --- Backtesting Results of Factors 1. Volatility Factor - Weekly excess return contribution: -0.06% (Fund Heavy Index), -0.74% (China Dividend Index), +0.35% (Micro-Cap Index)[2] - Annual excess return contribution: +1.64% (Fund Heavy Index), -8.90% (China Dividend Index), +10.20% (Micro-Cap Index)[2] - Weekly excess risk contribution: 13.90% (Fund Heavy Index), 79.76% (China Dividend Index), 9.09% (Micro-Cap Index)[2] - Annual excess risk contribution: 12.34% (Fund Heavy Index), 63.26% (China Dividend Index), 4.89% (Micro-Cap Index)[2] 2. Size Factor - Weekly excess return contribution: -0.50% (Fund Heavy Index), +2.57% (Micro-Cap Index)[2] - Annual excess return contribution: +2.16% (Fund Heavy Index), -11.14% (Micro-Cap Index)[2] - Weekly excess risk contribution: 55.79% (Fund Heavy Index), 83.84% (Micro-Cap Index)[2] - Annual excess risk contribution: 51.88% (Fund Heavy Index), 90.38% (Micro-Cap Index)[2] 3. Liquidity Factor - Weekly excess return contribution: +0.24% (China Dividend Index)[2] 4. Dividend Yield Factor - Weekly excess return contribution: +0.24% (China Dividend Index)[2] - Annual excess return contribution: +5.62% (China Dividend Index)[2] - Weekly excess risk contribution: 7.29% (China Dividend Index)[2] - Annual excess risk contribution: 8.57% (China Dividend Index)[2]