招商证券定量研究2026年度十大展望

Quantitative Models and Construction Methods Model 1: Active Quantitative Stock Selection Based on Free Cash Flow - Model Name: Active Quantitative Stock Selection Based on Free Cash Flow - Model Construction Idea: The model aims to select stocks with high free cash flow quality and enhance the selection with valuation, quality, dividend, and momentum factors[27] - Model Construction Process: 1. Sample Space: Exclude newly listed stocks (less than one year), ST and *ST stocks, and stocks in the comprehensive financial, banking, non-bank financial, and real estate sectors[29] 2. Initial Screening: Retain stocks with positive free cash flow, positive enterprise value, and positive net cash flow from operating activities over the past five years. Exclude stocks in the bottom 20% of profitability quality[29] 3. Free Cash Flow Selection: Construct free cash flow factors from valuation, quality, and growth dimensions. Neutralize these factors by market value and industry, then combine them equally to form a comprehensive free cash flow factor. Select the top 50% of stocks based on this factor to form a self-built cash flow stock pool[29] 4. Enhancement Dimensions: Introduce valuation, quality, dividend, and momentum factors to further enhance the stock pool[29] - Model Evaluation: The model has shown stable performance with an annualized return of 32.28% and an annualized excess return of 26.68% relative to the CSI 500, with an information ratio (IR) of 2.42[30] Model 2: High Dividend Stock Selection Strategy - Model Name: High Dividend Stock Selection Strategy - Model Construction Idea: The model focuses on selecting stocks with high dividend yields and stable dividend payments to construct a high dividend investment strategy[33] - Model Construction Process: 1. High Dividend Base Stock Pool: Select the top 20% of companies in each CITIC first-level industry based on dividend yield[35] 2. Avoid Dividend Trap: Select companies with an average dividend yield greater than 2% over the past three years and a standard deviation of dividend yield less than 2%[35] 3. Avoid Low Valuation Trap: Select companies with the current quarterly ROE greater than or equal to the same quarter last year and with a positive consensus forecast for future compound growth rate[35] 4. High Dividend Yield Portfolio: Select the top 30 companies based on dividend yield from the remaining stock pool[35] - Model Evaluation: The model has achieved an annualized excess return of 16.42% relative to the CSI Dividend Index, with an IR of 2.42[37] Model 3: Technical Growth Expectation Stock Selection Strategy - Model Name: Technical Growth Expectation Stock Selection Strategy - Model Construction Idea: The model aims to identify high-growth opportunities by constructing a future growth portfolio based on current growth indicators and enhancing it with technical factors[39] - Model Construction Process: 1. Growth Expectation Portfolio Construction: Select companies with the latest quarterly net profit growth to form the base stock pool. Exclude companies in the bottom 20% of quarterly ROE and select the top 50% based on the slope of quarterly ROE[41] 2. Technical Enhancement: Select the top 30% of companies based on the standardized unexpected earnings (SUE) indicator and the top 100 stocks based on excess returns on the day after earnings announcements. Further select the top 30 stocks based on the standard deviation of turnover rate moving average[41] - Model Evaluation: The model has shown an annualized return of 40% and an annualized excess return of 32.13% relative to the CSI 500, with an IR of 2.91[42] Model Backtest Results Active Quantitative Stock Selection Based on Free Cash Flow - Annualized Return: 32.28% - Annualized Excess Return: 26.68% - Information Ratio (IR): 2.42[30] High Dividend Stock Selection Strategy - Annualized Excess Return: 16.42% - Information Ratio (IR): 2.42[37] Technical Growth Expectation Stock Selection Strategy - Annualized Return: 40% - Annualized Excess Return: 32.13% - Information Ratio (IR): 2.91[42] Quantitative Factors and Construction Methods Factor 1: Free Cash Flow Factor - Factor Name: Free Cash Flow Factor - Factor Construction Idea: The factor aims to capture the quality of free cash flow from valuation, quality, and growth dimensions[29] - Factor Construction Process: 1. Valuation Dimension: Free Cash Flow to Firm/Enterprise Value (FCFF/EV) 2. Quality Dimension: Free Cash Flow to Firm/EBITDA (FCFF/EBITDA) 3. Growth Dimension: Free Cash Flow Growth Rate (FCFF Growth Rate) 4. Combination: Neutralize these factors by market value and industry, then combine them equally to form a comprehensive free cash flow factor[29] - Factor Evaluation: The factor has shown long-term effectiveness and stability in predicting future dividends and achieving excess returns[27] Factor Backtest Results Free Cash Flow Factor - Annualized Return: 32.28% - Annualized Excess Return: 26.68% - Information Ratio (IR): 2.42[30]