股指分红点位监控周报：7月合约即将到期，IC及IM合约持续深贴水-20250715

Quantitative Models and Factors Summary Quantitative Models and Construction Methods Model 1: Index Dividend Point Estimation Model - Model Name: Index Dividend Point Estimation Model - Model Construction Idea: The model aims to estimate the dividend points of index components to accurately calculate the futures basis and premium/discount levels by considering the impact of dividends on index points[12][39] - Model Construction Process: 1. Component Stock Weight: Adjust the weight of component stocks from the last disclosed date to the current date using the formula: $W_{n,t} = \frac{w_{i0} \times (1 + r_{n})}{\sum_{i=1}^{N} w_{i0} \times (1 + r_{n})}$ where $w_{i0}$ is the weight on the last disclosed date, and $r_{n}$ is the non-adjusted return of the stock[45] 2. Net Profit Prediction: Use historical net profit distribution to dynamically predict the annual net profit of the company. If the company has disclosed annual reports or performance forecasts, use the disclosed data; otherwise, estimate based on historical data[47][50] 3. Dividend Payout Ratio Prediction: Use historical dividend payout ratios to estimate the current year's payout ratio. If the company has not disclosed the dividend amount, estimate it using the predicted net profit and historical payout ratio[51] 4. Ex-Dividend Date Prediction: Predict the ex-dividend date based on historical intervals between announcement and ex-dividend dates using linear extrapolation[55] 5. Dividend Points Calculation: Calculate the dividend points using the formula: $\text{Dividend Points} = \sum_{n=1}^{N} \left( \frac{\text{Dividend Amount}}{\text{Total Market Value}} \times \text{Component Weight} \times \text{Index Closing Price} \right)$ where the ex-dividend date should be greater than the current date and less than or equal to the futures contract expiration date[39] - Model Evaluation: The model shows high accuracy in predicting dividend points for major indices like the SSE 50 and CSI 300, with minor deviations for the CSI 500 index[57][61] Model Backtesting Results - Index Dividend Point Estimation Model: - SSE 50 Index: Prediction error within 5 points[61] - CSI 300 Index: Prediction error within 5 points[61] - CSI 500 Index: Prediction error within 10 points[61] Quantitative Factors and Construction Methods Factor 1: Realized Dividend Yield - Factor Name: Realized Dividend Yield - Factor Construction Idea: Calculate the realized dividend yield of index components to monitor the dividend progress and remaining yield for the year[3][17] - Factor Construction Process: 1. Realized Dividend Yield Calculation: $\text{Realized Dividend Yield} = \sum_{i=1}^{N_{1}} \left( \frac{\text{Dividend Amount}}{\text{Total Market Value}} \times \text{Component Weight} \right)$ where $N_{1}$ is the number of companies that have paid dividends[17] 2. Remaining Dividend Yield Calculation: $\text{Remaining Dividend Yield} = \sum_{i=1}^{N_{2}} \left( \frac{\text{Dividend Amount}}{\text{Total Market Value}} \times \text{Component Weight} \right)$ where $N_{2}$ is the number of companies that have not yet paid dividends[17] - Factor Evaluation: The factor provides a clear view of the dividend progress and remaining potential dividends for the year, aiding in investment decisions[3][17] Factor Backtesting Results - Realized Dividend Yield: - SSE 50 Index: Realized yield 1.80%, remaining yield 0.46%[3] - CSI 300 Index: Realized yield 1.35%, remaining yield 0.53%[3] - CSI 500 Index: Realized yield 1.09%, remaining yield 0.20%[3] - CSI 1000 Index: Realized yield 0.83%, remaining yield 0.12%[3]