微盘股指数周报：短期上涨动能枯竭，控制仓位做好防御-20250616

Quantitative Models and Construction Methods 1. Model Name: Diffusion Index Model - Model Construction Idea: The diffusion index is used to monitor the critical points of market changes and predict potential turning points in the micro-cap stock index[5][35] - Model Construction Process: The diffusion index is calculated based on the relative price changes of constituent stocks over a specific time window. For example, if the horizontal axis is 0.95 and the vertical axis is 15 days, a value of 0.19 indicates that if all constituent stocks drop by 5% after 5 days, the diffusion index value will be 0.19. The formula and methodology are not explicitly provided but rely on historical and predictive data of constituent stocks[35][37] - Model Evaluation: The model is effective in identifying market turning points and provides actionable signals for left-side and right-side trading strategies[5][36] 2. Model Name: First Threshold Method (Left-Side Trading) - Model Construction Idea: This method triggers a sell signal when the diffusion index reaches a predefined threshold, indicating a left-side trading opportunity[5][39] - Model Construction Process: The first threshold method triggered a sell signal on May 8, 2025, when the diffusion index reached 0.9850[39] - Model Evaluation: It is suitable for early identification of market risks but may lead to premature signals in volatile markets[5][39] 3. Model Name: Delayed Threshold Method (Right-Side Trading) - Model Construction Idea: This method delays the sell signal until the diffusion index reaches a lower threshold, providing a more conservative approach[5][41] - Model Construction Process: The delayed threshold method triggered a sell signal on May 15, 2025, when the diffusion index reached 0.8975[43] - Model Evaluation: It reduces the risk of false signals but may miss early opportunities[5][43] 4. Model Name: Dual Moving Average Method (Adaptive Trading) - Model Construction Idea: This method uses two moving averages to adapt to market trends and trigger trading signals[5][44] - Model Construction Process: The dual moving average method triggered a sell signal on June 11, 2025[44] - Model Evaluation: It balances between early and delayed signals, making it suitable for dynamic market conditions[5][44] --- Model Backtesting Results 1. Diffusion Index Model - Current value: 0.70[35] - Predicted low point: Next Wednesday, with a potential cumulative decline of over 6%[5][36] 2. First Threshold Method - Triggered sell signal: May 8, 2025, at 0.9850[39] 3. Delayed Threshold Method - Triggered sell signal: May 15, 2025, at 0.8975[43] 4. Dual Moving Average Method - Triggered sell signal: June 11, 2025[44] --- Quantitative Factors and Construction Methods 1. Factor Name: PE_TTM Inverse Factor - Factor Construction Idea: Measures the inverse of the price-to-earnings ratio based on trailing twelve months (TTM) earnings[4][15] - Factor Construction Process: Rank IC for the week: 0.187 (historical average: 0.018)[4][15] - Factor Evaluation: Demonstrates strong predictive power for the week, significantly outperforming its historical average[4][15] 2. Factor Name: Profitability Factor - Factor Construction Idea: Evaluates the profitability of constituent stocks[4][15] - Factor Construction Process: Rank IC for the week: 0.177 (historical average: 0.023)[4][15] - Factor Evaluation: Shows robust performance, exceeding historical averages[4][15] 3. Factor Name: Unadjusted Stock Price Factor - Factor Construction Idea: Tracks the raw stock price without adjustments for splits or dividends[4][15] - Factor Construction Process: Rank IC for the week: 0.13 (historical average: -0.017)[4][15] - Factor Evaluation: Positive performance, reversing its historical negative trend[4][15] 4. Factor Name: Dividend Yield Factor - Factor Construction Idea: Measures the dividend yield of stocks[4][15] - Factor Construction Process: Rank IC for the week: 0.118 (historical average: 0.022)[4][15] - Factor Evaluation: Consistently outperforms its historical average[4][15] 5. Factor Name: PB Inverse Factor - Factor Construction Idea: Measures the inverse of the price-to-book ratio[4][15] - Factor Construction Process: Rank IC for the week: 0.116 (historical average: 0.034)[4][15] - Factor Evaluation: Strong weekly performance, well above historical norms[4][15] --- Factor Backtesting Results Top 5 Factors (Weekly Rank IC) 1. PE_TTM Inverse Factor: 0.187[4][15] 2. Profitability Factor: 0.177[4][15] 3. Unadjusted Stock Price Factor: 0.13[4][15] 4. Dividend Yield Factor: 0.118[4][15] 5. PB Inverse Factor: 0.116[4][15] Bottom 5 Factors (Weekly Rank IC) 1. 1-Year Volatility Factor: -0.188[4][15] 2. 10-Day Free Float Turnover Factor: -0.166[4][15] 3. Residual Volatility Factor: -0.143[4][15] 4. Trading Volume Factor: -0.142[4][15] 5. 10-Day Total Market Cap Turnover Factor: -0.139[4][15]