Report Summary 1. Report Industry Investment Rating No relevant content provided. 2. Core Viewpoints - Since 2020, the annualized hedging costs of stock index futures for the three major stock indices (SSE 50, CSI 300, and CSI 500) have been more significantly affected by stock dividends, especially on the expiration date [2]. - The hedging costs of the far - season contracts of SSE 50 and CSI 300 stock index futures have long been stable around zero, and are slightly higher than the break - even point in most periods. The cost of the far - season contracts of CSI 500 stock index futures also fluctuates around zero, and before 2023, it was overall superior to the contracts of SSE 50 and CSI 300 in terms of return performance, showing relatively better cost - return characteristics [2]. - Based on the prediction for the fourth quarter of 2025, under the current market structure, preferentially allocating hedging tools represented by IC stock index futures may be a better choice for constructing market - neutral strategies [2]. 3. Summary According to the Table of Contents 3.1. Introduction - In the practice of the stock market - neutral strategy, although the hedging means of the strategy portfolio have been significantly enriched, stock index futures are still the core hedging tool for constructing market - neutral exposure. The basis structure of stock index futures directly determines the hedging cost of the neutral strategy and affects the final return performance [6]. - The basis of stock index futures can be decomposed into three core driving dimensions: the cost dimension from the time value of funds, the cash - flow dimension from index component stock dividends during the period, and the sentiment and expectation dimension reflecting the balance of market long - and short - side forces. The model is simplified to: Futures price - Index price = Corrected basis - Index dividends during the period [6]. 3.2. Dividend Situations of the Three Major Index Component Stocks 3.2.1. Dividend Point Indices of the Three Major Stock Indices - The dividend behavior of the three major index component stocks has significant seasonal characteristics, with dividend payments highly concentrated from April to September each year, peaking from June to August, especially from June to July [8][12]. - Compared with the market practice before 2020, in recent years, the phenomenon of the three major index component stocks paying dividends in the fourth quarter has increased. Since 2023, the A - share market dividend pattern has shown three new trends: year - end dividends, postponed dividend dates for some companies, and a deeper impact of dividend behavior on stock index and derivatives pricing [9][12]. 3.2.2. Dividend Yield Situations of the Three Major Stock Indices - The dividend yields of the SSE 50 and CSI 300 indices showed a "V - shaped" trend of first decreasing and then increasing from 2020 to 2024, which is related to the market adjustment from the end of 2023 to the beginning of 2024. The dividend yield center of the CSI 500 index has shifted down compared with the level before 2020 [13][15]. - The average annual dividend yields of the SSE 50 and CSI 300 indices, representing large - cap blue - chip stocks, are stable in the range of 2% - 3%, while the average dividend yield of the CSI 500 index, representing small - and medium - cap stocks, is relatively low. The dividends from June to July have a significant impact on futures pricing and basis structure [15][16]. 3.3. Annualized Hedging Costs of the Three Major Stock Index Futures 3.3.1. Estimation of Historical Data of Annualized Hedging Costs of the Three Major Stock Index Futures - A simplified model is used to estimate the dividend points of index component stocks and calculate the corrected basis. The annualized hedging costs of the near - month, far - month, near - season, and far - season contracts of the three major stock index futures in the past three years are calculated [18][20]. - The hedging costs of the far - season contracts of SSE 50 and CSI 300 stock index futures are long - term stable around zero and slightly higher than the break - even point, with low historical average hedging costs. Before 2023, the far - season contracts of CSI 500 stock index futures were overall superior to those of SSE 50 and CSI 300 in terms of return performance [20]. - The annualized hedging cost of near - month contracts may show significant peaks, indicating that the basis of stock index futures contracts may fluctuate extremely (deep premium or discount) when approaching the expiration date, which affects the actual cost of roll - over operations and strategy returns [26]. 3.3.2. Prediction of the Performance of Annualized Hedging Costs of the Three Major Stock Index Futures in the Fourth Quarter - From the fourth quarter of 2025 to the beginning of 2026, the impact of dividends on the basis of stock index futures and hedging strategies has weakened. The hedging costs of the current IC and IF main contracts are generally positive, providing a favorable window for market - neutral strategies [27][28]. - Based on the closing data on September 22, 2025, the overall hedging costs of the three major stock index futures are relatively low. After considering dividends, most of the hedging costs of IC and IF contracts are positive, especially for IC near - month and far - month contracts. The hedging costs of IH contracts are relatively high and even negative after considering dividends [29][30]. - Currently, market - neutral strategies using IC or IF futures contracts for hedging have relative advantages. IC near - season main contracts have lower hedging costs, while IF far - season contracts have more obvious cost advantages. IH contracts have relatively low cost - performance. Therefore, preferentially allocating hedging tools represented by IC stock index futures may be a better choice [30][31].

Quantitative Models and Construction Methods - **Model Name**: Dividend Forecast Model **Model Construction Idea**: The model aims to predict the impact of dividends on index futures pricing by estimating the dividend distribution of index constituent stocks and its influence on futures contracts[9][20][23] **Model Construction Process**: 1. **Estimate Net Profit of Constituent Stocks**: Use annual reports, earnings forecasts, and other financial data to estimate the net profit of each constituent stock[21][23] 2. **Calculate Pre-Tax Total Dividends**: Based on the assumption that the dividend payout ratio remains constant, calculate the total pre-tax dividends for each stock[24][27] 3. **Assess Dividend Impact on Index**: - Dividend Yield = Total Post-Tax Dividends / Latest Market Value - Dividend Points = Stock Weight × Dividend Yield - Adjust stock weights using the formula: $$\mathrm{w_{it}={\frac{w_{i0}\times\mathrm{\(\1+R\)}}{\sum_{1}^{n}w_{i0}\times\mathrm{\(\1+R\)}}}}$$ where \(w_{i0}\) is the initial weight, and \(R\) is the stock's return over the period[24] 4. **Predict Dividend Impact on Futures Contracts**: - Estimate ex-dividend dates based on historical patterns or announced schedules - Aggregate all dividends before the contract's settlement date to calculate the total impact on futures pricing[25][26][28] **Model Evaluation**: The model provides a systematic approach to quantify dividend impacts, but its accuracy depends on the reliability of assumptions and historical data[9][20][23] - **Model Name**: Futures Pricing Model with Discrete Dividends **Model Construction Idea**: This model calculates the theoretical price of index futures by incorporating the present value of discrete dividend distributions during the contract period[29] **Model Construction Process**: - Formula: $$F_t = (S_t - D)(1 + r)$$ where \(F_t\) is the futures price, \(S_t\) is the spot price, \(D\) is the present value of dividends, and \(r\) is the risk-free rate[29] **Model Evaluation**: The model is effective for scenarios with distinct dividend distributions but may not capture continuous dividend flows accurately[29] - **Model Name**: Futures Pricing Model with Continuous Dividends **Model Construction Idea**: This model assumes dividends are distributed continuously over time and calculates the theoretical futures price accordingly[30] **Model Construction Process**: - Formula: $$F_t = S_t e^{(r-d)(T-t)}$$ where \(F_t\) is the futures price, \(S_t\) is the spot price, \(r\) is the risk-free rate, \(d\) is the annualized dividend yield, and \(T-t\) is the time to maturity[30] **Model Evaluation**: Suitable for markets with frequent and evenly distributed dividends, providing a more realistic pricing framework in such scenarios[30] Model Backtesting Results - **Dividend Forecast Model**: - Remaining dividend impact on July contracts: IH (0.82%), IF (0.57%), IC (0.16%), IM (0.14%)[15] - Annualized hedging costs (excluding dividends, 365-day basis): IH (-4.89%), IF (-2.72%), IC (12.34%), IM (16.97%)[6][10][12][13][14] - **Futures Pricing Model with Discrete Dividends**: - Not explicitly tested in the report - **Futures Pricing Model with Continuous Dividends**: - Not explicitly tested in the report Quantitative Factors and Construction Methods - **Factor Name**: Dividend Yield Factor **Factor Construction Idea**: Measures the dividend yield of index constituent stocks to assess their contribution to the overall index dividend impact[24] **Factor Construction Process**: - Formula: Dividend Yield = Total Post-Tax Dividends / Latest Market Value[24] **Factor Evaluation**: Provides a direct measure of dividend contribution but may be sensitive to market value fluctuations[24] - **Factor Name**: Stock Weight Adjustment Factor **Factor Construction Idea**: Adjusts the weight of each stock in the index based on its return over a specified period[24] **Factor Construction Process**: - Formula: $$\mathrm{w_{it}={\frac{w_{i0}\times\mathrm{\(\1+R\)}}{\sum_{1}^{n}w_{i0}\times\mathrm{\(\1+R\)}}}}$$ where \(w_{i0}\) is the initial weight, and \(R\) is the stock's return over the period[24] **Factor Evaluation**: Enhances the accuracy of dividend impact calculations by accounting for stock performance dynamics[24] Factor Backtesting Results - **Dividend Yield Factor**: - Not explicitly tested in the report - **Stock Weight Adjustment Factor**: - Not explicitly tested in the report

- The report discusses the impact of dividends on stock index futures, specifically for the contracts of the SSE 50, CSI 300, CSI 500, and CSI 1000 indices [1][2][3][4] - The latest dividend forecast model predicts the dividend points for the June contracts of the SSE 50, CSI 300, CSI 500, and CSI 1000 indices to be 12.10, 16.30, 18.75, and 17.78, respectively [8][11] - The annualized hedging costs (excluding dividends, calculated on a 365-day basis) for the June contracts of the SSE 50, CSI 300, CSI 500, and CSI 1000 indices are 3.05%, 1.54%, 8.11%, and 14.77%, respectively [8][11] - The report provides detailed data on the closing prices, dividend points, actual spreads, and dividend-inclusive spreads for the June, July, September, and December contracts of the SSE 50, CSI 300, CSI 500, and CSI 1000 indices [2][3][4] - The remaining impact of dividends on the June contracts of the SSE 50, CSI 300, CSI 500, and CSI 1000 indices is 0.45%, 0.42%, 0.33%, and 0.29%, respectively [12][13][14][15][16] - The report outlines the process for predicting dividends, which includes estimating the net profit of constituent stocks, calculating the total pre-tax dividends for each stock, and determining the impact of dividends on the index and each contract [9][21][24][25][26][27][28][29][30] - The theoretical pricing model for stock index futures is provided, including formulas for both discrete and continuous dividend distributions [32][33] Model and Factor Construction - **Model Name**: Dividend Impact Prediction Model - **Model Construction Idea**: The model aims to predict the impact of dividends on stock index futures contracts by estimating the dividends of constituent stocks and calculating their effect on the index and futures contracts [9][21] - **Model Construction Process**: 1. Estimate the net profit of constituent stocks using available information such as annual reports, quick reports, warnings, and analyst forecasts [23][24] 2. Calculate the total pre-tax dividends for each stock based on the estimated net profit and historical dividend rates [25][28] 3. Determine the impact of dividends on the index by calculating the dividend yield and the dividend points for each stock [26] 4. Estimate the ex-dividend dates and calculate the theoretical impact of dividends on each futures contract [27][29][30] 5. Use the theoretical pricing model for stock index futures to incorporate the impact of dividends into the futures prices [32][33] - **Model Evaluation**: The model provides a systematic approach to predict the impact of dividends on stock index futures, considering various factors such as net profit estimates, dividend rates, and ex-dividend dates [9][21][24] Model Backtesting Results - **SSE 50 Futures (June Contract)**: - Closing Price: 2673.60 - Dividend Points: 12.10 - Actual Spread: -15.25 - Dividend-Inclusive Spread: -3.15 - Remaining Impact: 0.45% - Annualized Hedging Cost (365 days): 3.05% - Annualized Hedging Cost (243 days): 2.84% [2][12] - **CSI 300 Futures (June Contract)**: - Closing Price: 3855.40 - Dividend Points: 16.30 - Actual Spread: -18.58 - Dividend-Inclusive Spread: -2.28 - Remaining Impact: 0.42% - Annualized Hedging Cost (365 days): 1.54% - Annualized Hedging Cost (243 days): 1.43% [2][13] - **CSI 500 Futures (June Contract)**: - Closing Price: 5725.40 - Dividend Points: 18.75 - Actual Spread: -36.68 - Dividend-Inclusive Spread: -17.93 - Remaining Impact: 0.33% - Annualized Hedging Cost (365 days): 8.11% - Annualized Hedging Cost (243 days): 7.56% [3][14] - **CSI 1000 Futures (June Contract)**: - Closing Price: 6100.20 - Dividend Points: 17.78 - Actual Spread: -52.64 - Dividend-Inclusive Spread: -34.87 - Remaining Impact: 0.29% - Annualized Hedging Cost (365 days): 14.77% - Annualized Hedging Cost (243 days): 13.77% [4][15]