股指分红点位监控周报：年度分红预测开启，各股指期货主力合约均深度贴水-20250410

Quantitative Models and Construction Methods Dividend Points Estimation Model - Model Name: Dividend Points Estimation Model - Model Construction Idea: This model estimates the dividend points of stock indices to account for the impact of constituent stock dividends on index futures pricing. It is essential for accurately calculating the basis and premium/discount levels of index futures contracts[12][39]. - Model Construction Process: 1. Formula: The dividend points during the period from the current date $t$ to the futures contract expiration date $T$ are calculated as: $\text{Dividend Points} = \sum_{n=1}^{N} \frac{\text{Dividend Amount of Stock } n}{\text{Total Market Value of Stock } n} \times \text{Weight of Stock } n \times \text{Index Closing Price}$ - $N$: Number of constituent stocks in the index - $t$: Current date - $T$: Futures contract expiration date - $\text{Dividend Amount of Stock } n$: Dividend amount of the $n$-th stock - $\text{Weight of Stock } n$: Weight of the $n$-th stock in the index - $\text{Index Closing Price}$: Closing price of the index[39] 2. Steps: - Identify constituent stocks and their weights - Estimate dividend amounts for stocks without announced dividends using the formula: $\text{Dividend Amount} = \text{Net Profit} \times \text{Dividend Payout Ratio}$ - Net profit is predicted using historical profit distribution patterns - Dividend payout ratio is estimated using historical averages[48][52] - Predict ex-dividend dates using historical intervals and linear extrapolation[52][57] - Model Evaluation: The model demonstrates high accuracy for indices like the SSE 50 and CSI 300, with prediction errors generally within 5 points. For the CSI 500, the error is slightly larger but remains stable around 10 points[62]. --- Quantitative Factors and Construction Methods Constituent Stock Weight Adjustment Factor - Factor Name: Constituent Stock Weight Adjustment Factor - Factor Construction Idea: Adjusts the weights of index constituent stocks to reflect daily changes due to price fluctuations, ensuring more precise dividend point calculations[46]. - Factor Construction Process: 1. Formula: $W_{n,t} = \frac{w_{n,0} \times (1 + r_{n})}{\sum_{i=1}^{N} w_{i,0} \times (1 + r_{i})}$ - $W_{n,t}$: Adjusted weight of stock $n$ on day $t$ - $w_{n,0}$: Weight of stock $n$ on the last disclosed date - $r_{n}$: Non-split-adjusted return of stock $n$ from the last disclosed date to day $t$ - $N$: Total number of constituent stocks[46] 2. Steps: - Use the most recent disclosed weights as the base - Adjust weights daily based on stock price movements - Incorporate daily weight data from the China Securities Index Company for higher accuracy[47] - Factor Evaluation: This adjustment method ensures precise daily weight calculations, reducing errors caused by stock adjustments, unlocks, or rights issues[47]. Net Profit Prediction Factor - Factor Name: Net Profit Prediction Factor - Factor Construction Idea: Predicts annual net profit for constituent stocks based on historical profit distribution patterns, enabling accurate dividend amount estimation[48]. - Factor Construction Process: 1. Classify companies into two categories: - Stable Profit Distribution: Predict using historical profit distribution patterns - Unstable Profit Distribution: Use the previous year's profit as the prediction[51] 2. For companies with disclosed financial reports, use the reported net profit. For those with performance forecasts, use the average of the forecast range[51]. - Factor Evaluation: The dynamic prediction method effectively captures the profit trends of stable companies, while the fallback approach ensures reasonable estimates for volatile companies[51]. Dividend Payout Ratio Prediction Factor - Factor Name: Dividend Payout Ratio Prediction Factor - Factor Construction Idea: Estimates the dividend payout ratio using historical averages, assuming stability in payout policies for most companies[52]. - Factor Construction Process: 1. Steps: - If the company paid dividends last year, use the previous year's payout ratio - If no dividends were paid last year, use the average payout ratio of the past three years - If the company has never paid dividends, assume no dividends for the current year - Cap the payout ratio at 100% to avoid unrealistic estimates[54] - Factor Evaluation: The method leverages historical stability in payout ratios for reliable predictions, with adjustments for outliers[54]. Ex-Dividend Date Prediction Factor - Factor Name: Ex-Dividend Date Prediction Factor - Factor Construction Idea: Predicts ex-dividend dates using historical intervals and default dates for companies without sufficient data[52]. - Factor Construction Process: 1. Steps: - Use announced ex-dividend dates if available - For companies with historical data, calculate the average interval between announcement and ex-dividend dates, and apply linear extrapolation - For companies without historical data, use default dates based on typical market behavior (e.g., July 31, August 31, or September 30)[57] - Factor Evaluation: The method achieves high accuracy for most companies, with over 90% of predictions aligning with actual dates[57]. --- Backtesting Results of Models and Factors Dividend Points Estimation Model - Accuracy: - SSE 50 Index: Prediction error within 5 points - CSI 300 Index: Prediction error within 5 points - CSI 500 Index: Prediction error within 10 points[62] - Futures Contracts: - SSE 50 Futures: High prediction accuracy for dividend points - CSI 300 Futures: High prediction accuracy for dividend points - CSI 500 Futures: Slightly larger deviations but still acceptable[62]