Quantitative Models and Construction Methods - **Model Name**: Dividend Style Timing Model **Model Construction Idea**: The model is designed to evaluate the timing of dividend style investments by aggregating multiple factors into a composite score. The score reflects the attractiveness of the dividend style based on various market and macroeconomic indicators[6][9]. **Model Construction Process**: 1. The model aggregates multiple factors, including macroeconomic indicators, market sentiment, and momentum factors, to calculate a composite factor value. 2. The composite factor value is updated weekly to reflect the latest market conditions. 3. The composite factor value is used to determine whether the dividend style is in a favorable or unfavorable state. 4. For the week of March 16, 2026, to March 20, 2026, the composite factor value was -0.26, which remained negative but showed improvement compared to the previous week's value of -0.39[6][9]. **Model Evaluation**: The model effectively integrates multiple factors to provide a comprehensive view of the dividend style's timing. However, the composite factor value has not yet signaled a bullish outlook[6][9]. Quantitative Factors and Construction Methods 1. **Factor Name**: Sentiment Factor **Factor Construction Idea**: Measures market sentiment to assess its impact on the dividend style. Positive sentiment contributes positively to the composite score[9]. **Factor Construction Process**: - Sentiment is quantified using market data and behavioral indicators. - The factor's contribution to the composite score is calculated based on its correlation with dividend style performance. **Factor Evaluation**: The sentiment factor positively contributed to the composite score during the observed period, indicating its relevance in the model[9]. 2. **Factor Name**: Momentum Factor **Factor Construction Idea**: Captures the momentum of the dividend style to predict its future performance. Positive momentum contributes positively to the composite score[9]. **Factor Construction Process**: - Momentum is calculated using historical price data of dividend-related assets. - The factor's contribution is determined by its alignment with the overall market trend. **Factor Evaluation**: The momentum factor positively contributed to the composite score, supporting the dividend style's relative strength during the period[9]. 3. **Factor Name**: U.S. 10-Year Treasury Yield **Factor Construction Idea**: Reflects the impact of U.S. Treasury yields on the dividend style. Higher yields typically exert downward pressure on dividend-related assets[9][12]. **Factor Construction Process**: - The 10-year Treasury yield is directly incorporated as a factor. - Its contribution is calculated based on its inverse relationship with dividend style performance. **Factor Evaluation**: The factor's negative contribution to the composite score weakened during the observed period, as U.S. Treasury yields rose to a six-month high, reducing their suppressive effect on the dividend style[9][12]. 4. **Factor Name**: Dividend Relative Net Value **Factor Construction Idea**: Measures the relative performance of dividend-related assets compared to a benchmark. Positive relative performance contributes positively to the composite score[12]. **Factor Construction Process**: - The relative net value is calculated by comparing the performance of dividend-related assets to a benchmark index. - The factor's contribution is determined by its deviation from the benchmark. **Factor Evaluation**: This factor made a strong positive contribution to the composite score, reflecting the dividend style's outperformance during the period[12]. 5. **Factor Name**: Financing Net Purchase **Factor Construction Idea**: Tracks net financing purchases to gauge investor interest in dividend-related assets. Positive net purchases contribute positively to the composite score[12]. **Factor Construction Process**: - Net financing purchases are calculated using market transaction data. - The factor's contribution is based on its correlation with dividend style performance. **Factor Evaluation**: This factor showed a positive contribution during the observed period, indicating increased investor interest in dividend-related assets[12]. Model Backtesting Results - **Dividend Style Timing Model**: - Composite Factor Value (March 20, 2026): -0.26 - Composite Factor Value (March 13, 2026): -0.39 - Composite Factor Value (February 27, 2026): -0.54[6][12] Factor Backtesting Results 1. **Sentiment Factor**: Positive contribution to the composite score[9] 2. **Momentum Factor**: Positive contribution to the composite score[9] 3. **U.S. 10-Year Treasury Yield**: Negative contribution, but the suppressive effect weakened during the period[9][12] 4. **Dividend Relative Net Value**: Strong positive contribution to the composite score[12] 5. **Financing Net Purchase**: Positive contribution to the composite score[12]

Investment Rating - The report maintains a negative investment rating for the dividend style timing model, with a composite factor value of -0.39 for the week of March 9 to March 13, 2026, which is consistent with the previous week's value of -0.38, indicating no bullish signal has been issued [1][6]. Core Insights - The recent performance of dividends has been relatively good, with the momentum factor contributing positively. However, other variables have shown little change and contributed negatively. The decline in U.S. Treasury yields and the recovery in industry prosperity have negatively impacted dividends, resulting in a sustained negative score due to the interplay of multiple factors [4][7]. Summary by Sections Model Latest Results - The composite factor value for the dividend style timing model for the week of March 9 to March 13, 2026, is -0.39, remaining negative and unchanged from the previous week [6][11]. Factor Contributions - Recent contributions from various factors include: - Momentum factor showing positive contribution - U.S. Treasury yield (10-year) at -0.81, contributing negatively - Industry average prosperity at 0.91, contributing positively - Dividend relative net value at 1.59, contributing positively - Other factors such as M2 growth and financing net purchases showing mixed contributions [12][7].

- The dividend timing model was constructed in the report "Dividend Style Timing Scheme" on October 16, 2025[6] - The comprehensive factor value of the dividend timing model for the week of March 9 to March 13, 2026, was -0.39, remaining negative and almost unchanged compared to the previous week (-0.38)[1][6] - Momentum factor contributed positively to the dividend performance, while other variables such as the decline in US Treasury yields and the recovery of industry sentiment contributed negatively, resulting in an overall negative score[4][7] - Factor values for March 13, 2026, include: - China Non-Manufacturing PMI (Service Industry): 0.14[12] - China M2 YoY: 0.31[12] - US 10-Year Treasury Yield: -0.81[12] - Dividend relative net value: 1.59[12] - CSI Dividend Stock Dividend Yield - 10-Year Treasury Yield: -0.22[12] - Financing net purchase: 0.74[12] - Industry average sentiment: 0.91[12]

Quantitative Models and Construction Methods 1. Model Name: Global Asset Allocation Model Based on Artificial Intelligence - **Model Construction Idea**: This model applies machine learning techniques to global asset allocation, leveraging factor investment principles to rank assets and construct a monthly quantitative equal-weight allocation strategy[38] - **Model Construction Process**: - The model evaluates various global assets using machine learning algorithms to assign scores to each asset - Based on these scores, the model determines the optimal allocation weights for each asset class - For March, the recommended weights were: Government Bond Index 75.99%, SHFE Gold 23.50%, and Nasdaq 0.51%[38][42] - **Model Evaluation**: The model demonstrates superior performance in terms of return, risk-adjusted return, and drawdown control compared to the benchmark[40] 2. Model Name: Stock-Bond Allocation Model Based on Dynamic Macro Event Factors - **Model Construction Idea**: This model integrates a macro timing module and a risk budgeting framework to allocate between stocks and bonds for different risk preferences (Conservative, Balanced, and Aggressive)[45] - **Model Construction Process**: - The macro timing module evaluates signals from economic growth and monetary liquidity dimensions - The risk budgeting framework adjusts stock and bond weights based on the strength of these signals - For February, the stock weights were 30.00% (Aggressive), 13.81% (Balanced), and 0.00% (Conservative), with the remaining allocated to bonds[45][46] - **Model Evaluation**: The model has shown strong historical performance, with high annualized returns and Sharpe ratios across all risk profiles[46][51] 3. Model Name: Dividend Style Timing Model - **Model Construction Idea**: This model uses a dynamic macro event factor system based on economic growth and monetary liquidity indicators to time allocations to the CSI Dividend Index[52] - **Model Construction Process**: - The model evaluates 10 indicators across economic growth and monetary liquidity dimensions - For March, the final composite signal was 1, with economic growth indicators (e.g., power generation, PPI YoY, PPI-CPI spread) signaling bullish, while no monetary liquidity indicators signaled bullish[52][54] - **Model Evaluation**: The strategy outperforms the CSI Dividend Total Return Index in terms of annualized return, volatility, and drawdown, with a smoother net value curve[52][53] --- Model Backtesting Results 1. Global Asset Allocation Model Based on Artificial Intelligence - **Annualized Return**: 16.48% - **Annualized Volatility**: 6.78% - **Maximum Drawdown**: -6.66% - **Sharpe Ratio**: 1.15 - **Year-to-Date Return**: 5.72%[40][43] 2. Stock-Bond Allocation Model Based on Dynamic Macro Event Factors - **Aggressive Profile**: - Annualized Return: 20.07% - Annualized Volatility: 14.00% - Maximum Drawdown: -13.72% - Sharpe Ratio: 1.31 - Year-to-Date Return: 5.42% - **Balanced Profile**: - Annualized Return: 10.79% - Annualized Volatility: 8.09% - Maximum Drawdown: -6.77% - Sharpe Ratio: 1.19 - Year-to-Date Return: 1.81% - **Conservative Profile**: - Annualized Return: 5.85% - Annualized Volatility: 3.19% - Maximum Drawdown: -3.55% - Sharpe Ratio: 1.49 - Year-to-Date Return: 0.83%[46][51] 3. Dividend Style Timing Model - **Annualized Return**: 15.85% - **Annualized Volatility**: 16.17% - **Maximum Drawdown**: -21.22% - **Sharpe Ratio**: 0.93 - **Recent 1-Month Return**: 1.35%[52][53]

Investment Rating - The report indicates a negative investment rating for the dividend style timing model, with a composite factor value of -0.38 for the week of March 2 to March 6, 2026, down from -0.10 the previous week, and no bullish signal was issued [1][5]. Core Insights - The primary marginal change affecting the model is the recent decline in U.S. Treasury yields, which has intensified its suppressive effect on dividends. Market sentiment has shown some recovery, contributing negatively to dividend excess, while other variable changes remain minor. Overall, the combination of these factors results in a negative score [8][11]. Summary by Relevant Sections - **Model Latest Results**: The composite factor value for the dividend style timing model is -0.38 for the week of March 2 to March 6, 2026, indicating a continued negative trend compared to the previous week's value of -0.10, with no bullish signal generated [5][11]. - **Factor Analysis**: The analysis shows that the largest marginal change is attributed to the recent decline in U.S. Treasury yields, which has a growing suppressive effect on dividends. Additionally, the recovery in industry sentiment has negatively impacted dividends. The overall score remains negative due to the interplay of these factors [8][11]. - **Individual Factor Values**: The report provides specific factor values as of March 6, 2026, including: - Non-manufacturing PMI for China: 0.14 - M2 YoY for China: 0.31 - U.S. 10-Year Treasury Yield: -0.70 - Relative net value of dividends: 0.76 - Dividend yield relative to 10-year government bond yield: -0.11 - Net financing purchases: 0.26 - Average industry sentiment: 0.97 [11].

- The dividend timing model was constructed in the report "Dividend Style Timing Scheme" on October 16, 2025[5] - The latest composite factor value of the dividend timing model for the week of 20260302 to 20260306 is -0.38, which remains negative and has declined compared to the previous week's value of -0.10[1][5] - The composite factor value reflects multiple influences, including the recent decline in US Treasury yields, which has strengthened its suppressive effect on dividends, and the recovery of market sentiment, which has negatively contributed to dividend excess returns[3][8] - The sub-factors of the model include variables such as China's non-manufacturing PMI (service sector), China's M2 year-on-year growth, US 10-year Treasury yield, relative net value of dividends, dividend yield spread over 10-year Chinese Treasury yield, net financing purchases, and industry average prosperity[11] - Factor values for 20260306 are as follows: China's non-manufacturing PMI (service sector) at 0.14, China's M2 year-on-year growth at 0.31, US 10-year Treasury yield at -0.70, relative net value of dividends at 0.76, dividend yield spread over 10-year Chinese Treasury yield at -0.11, net financing purchases at 0.26, and industry average prosperity at 0.97[11]

Investment Rating - The report indicates a slight negative signal with a comprehensive factor value of -0.10 for the dividend style timing model during the period from February 24 to February 27, 2026, after two consecutive weeks of positive values [6][7]. Core Insights - The recent decline in U.S. Treasury yields has increased its suppressive effect on dividends, while market sentiment has improved, leading to a reduction in the excess performance of dividends. This combination resulted in a slight decrease in the model score to a negative value [6][7]. - The model's factor values are fluctuating around the zero axis, suggesting a potential key point for style switching that warrants ongoing observation [6][7]. - The overall strong momentum of dividends and relatively low market sentiment contributed positively to the dividend style, while the decline in U.S. Treasury yields and the recovery in industry prosperity had a negative impact [7]. Summary by Relevant Sections Model Latest Results - The comprehensive factor value for the dividend style timing model was -0.10 for the week of February 24 to February 27, 2026, down from 0.09 the previous week [6][7]. - The model indicates that fluctuations around the zero axis are normal and may indicate a critical point for style switching [6]. Factor Insights - The largest marginal change was attributed to the recent decline in U.S. Treasury yields, which has intensified its negative impact on dividends [7]. - The individual factor values as of February 27, 2026, include: - Non-manufacturing PMI for China: -0.13 - M2 YoY for China: 0.21 - 10-Year U.S. Treasury Yield: -0.49 - Relative net value of dividends: 0.54 - Dividend yield of CSI Dividend Index minus 10-Year Treasury Yield: 0.03 - Net financing purchases: -1.59 - Average industry prosperity: 1.12 [11].

- The Dividend Style Timing Model's comprehensive factor value for the week of 20260224 to 20260227 was -0.10, turning slightly negative after being positive for two consecutive weeks. The previous week's factor value (20260209 to 20260213) was 0.09[1][4][6] - The factor value's slight fluctuation around the zero axis is normal, indicating a potential key point for style switching, which requires continuous observation to confirm successful switching[1][4][6] - The most significant marginal change was the recent decline in US Treasury yields, which increased the suppressive effect on dividends. Meanwhile, market sentiment improved, reducing the excess performance of dividends, leading to a slight decrease in the model score to negative[7] - The model's score was negatively impacted by the decline in US Treasury yields and the recovery in industry sentiment, while the low market sentiment and strong dividend momentum contributed positively to the dividend style[7] - The updated factor values for various components as of 20260227 are as follows: China Non-Manufacturing PMI: Services (-0.13), China M2 YoY (0.21), US 10-Year Treasury Yield (-0.49), Dividend Relative Net Value (0.54), CSI Dividend Yield - 10-Year China Bond Yield (0.03), Net Financing Purchase (-1.59), and Industry Average Prosperity (1.12)[11]

Investment Rating - The report indicates a positive investment rating for the dividend style timing model, with a composite factor value of 0.09, signaling a positive trend for two consecutive weeks [1][6]. Core Insights - The dividend style has shown good recent performance, with the momentum factor maintaining a positive contribution. The downward trend in U.S. Treasury yields continues to weaken, reducing pressure on dividends. Market sentiment has cooled, contributing positively to the dividend style. Multiple factors are at play, maintaining a positive score [4][7]. Summary by Sections Model Latest Results - The composite factor value for the dividend style timing model for the week of February 9 to February 13, 2026, is 0.09, consistent with the previous week [6][11]. Factor Analysis - The individual indicators are close to the previous week, with the following key factors: - Non-manufacturing PMI for China: -0.13 - M2 YoY for China: 0.21 - 10-Year U.S. Treasury Yield: -0.23 - Relative net value of dividends: 0.89 - Dividend yield of CSI Dividend Index minus 10-Year Treasury yield: 0.17 - Net financing purchases: -1.86 - Average industry prosperity: 1.02 - These factors collectively contribute to the positive score [12].

- The Dividend Style Timing Model's comprehensive factor value was 0.09 for the week of 20260209 to 20260213, maintaining a positive signal for two consecutive weeks[1][6] - The model's sub-indicators were similar to the previous week, with the momentum factor continuing to contribute positively and the downward trend in US Treasury rates weakening, reducing the pressure on dividends[7] - Market sentiment cooled, contributing positively to the dividend style, resulting in a maintained positive score due to multiple factors[7] Quantitative Models and Construction Methods 1. **Model Name**: Dividend Style Timing Model - **Model Construction Idea**: The model aims to time the dividend style by evaluating various factors that influence dividend performance[6] - **Model Construction Process**: The model aggregates multiple factors to generate a comprehensive factor value. These factors include momentum, US Treasury rates, market sentiment, and others. The comprehensive factor value is calculated as a weighted sum of these individual factors[6][7] - **Model Evaluation**: The model has shown consistent positive signals over the past two weeks, indicating its effectiveness in capturing the dividend style's performance[1][6] Model Backtesting Results 1. **Dividend Style Timing Model**: Comprehensive factor value of 0.09 for the week of 20260209 to 20260213, same as the previous week[1][6] Quantitative Factors and Construction Methods 1. **Factor Name**: Momentum Factor - **Factor Construction Idea**: Measures the recent performance trend of dividend stocks to predict future performance[7] - **Factor Construction Process**: The momentum factor is calculated based on the recent price movements of dividend stocks. The exact formula is not provided, but it typically involves calculating the rate of change or moving averages of stock prices[7] - **Factor Evaluation**: The momentum factor has maintained a positive contribution to the model, indicating its reliability in capturing performance trends[7] 2. **Factor Name**: US Treasury Rates - **Factor Construction Idea**: Evaluates the impact of US Treasury rates on dividend stocks[7] - **Factor Construction Process**: The factor is calculated based on the yield of 10-year US Treasury bonds. The exact formula is not provided, but it typically involves comparing the yield to historical averages or other benchmarks[7] - **Factor Evaluation**: The downward trend in US Treasury rates has weakened, reducing the pressure on dividend stocks and contributing positively to the model[7] 3. **Factor Name**: Market Sentiment - **Factor Construction Idea**: Assesses the overall market sentiment and its impact on dividend stocks[7] - **Factor Construction Process**: The factor is calculated based on various market sentiment indicators, such as investor surveys, trading volumes, and volatility indices. The exact formula is not provided[7] - **Factor Evaluation**: Market sentiment has cooled, contributing positively to the dividend style[7] Factor Backtesting Results 1. **Momentum Factor**: Positive contribution to the model[7] 2. **US Treasury Rates**: Reduced pressure on dividend stocks, contributing positively[7] 3. **Market Sentiment**: Positive contribution due to cooling market sentiment[7]