红利风格择时周报（0119-0123）-20260126

Quantitative Models and Construction Methods - Model Name: Dividend Style Timing Model Model Construction Idea: The model is designed to time the dividend style by aggregating multiple factors that influence the performance of dividend-related assets. The comprehensive factor value is used to determine the timing signal for the dividend style[6] Model Construction Process: 1. The model aggregates multiple factors, including financing net purchases, U.S. Treasury yields, market sentiment, and recent dividend performance, among others[7] 2. The comprehensive factor value is calculated as a weighted sum of these individual factors, with weights determined based on historical data and their predictive power for dividend style performance[6][7] Model Evaluation: The model provides a systematic approach to assess the timing of the dividend style, but its effectiveness depends on the stability of the relationships between the factors and the dividend style performance[6][7] Model Backtesting Results - Dividend Style Timing Model: The comprehensive factor value for the week of January 19, 2026, to January 23, 2026, was -0.57, an improvement from the previous week's value of -0.77. However, the value remained below 0, indicating no positive signal for the dividend style[6][7] Quantitative Factors and Construction Methods - Factor Name: U.S. 10-Year Treasury Yield Factor Construction Idea: This factor measures the impact of U.S. Treasury yields on dividend style performance, as lower yields typically favor dividend-paying stocks[7] Factor Construction Process: The factor value is derived from the yield of the 10-year U.S. Treasury bond. A negative contribution indicates that the yield's trend is suppressing the dividend style[7] - Factor Name: Financing Net Purchases Factor Construction Idea: This factor captures the influence of financing activities on the dividend style, with higher net purchases indicating stronger support for the style[7] Factor Construction Process: The factor value is calculated based on the net amount of financing purchases. A positive value indicates a supportive environment for the dividend style[7] - Factor Name: Market Sentiment Factor Construction Idea: This factor reflects the overall market sentiment, with higher sentiment levels potentially leading to a shift away from defensive styles like dividends[7] Factor Construction Process: The factor value is derived from sentiment indicators, such as market volatility and investor surveys. A high sentiment value suggests a potential headwind for the dividend style[7] - Factor Name: Dividend Relative Net Value Performance Factor Construction Idea: This factor measures the recent performance of dividend-related assets relative to the broader market, indicating the style's momentum[7] Factor Construction Process: The factor value is calculated as the relative performance of dividend-related indices compared to the market index. A negative value indicates underperformance[7] Factor Backtesting Results - U.S. 10-Year Treasury Yield: Factor value on January 23, 2026: -0.56; January 16, 2026: -0.71; December 31, 2025: -0.84[10] - Financing Net Purchases: Factor value on January 23, 2026: 1.86; January 16, 2026: 3.00; December 31, 2025: 0.82[10] - Market Sentiment: Factor value on January 23, 2026: 0.78; January 16, 2026: 1.05; December 31, 2025: 1.54[10] - Dividend Relative Net Value Performance: Factor value on January 23, 2026: -1.46; January 16, 2026: -1.71; December 31, 2025: -0.69[10]