主动量化周报：成交量回落触底，指数短期或将反弹

Quantitative Models and Construction Methods - Model Name: Oversold Rebound Signal Model Construction Idea: This model identifies potential short-term rebounds in industries that have experienced significant price declines, aiming to capture mean-reversion opportunities[14] Model Construction Process: The model uses historical data to identify oversold conditions based on predefined thresholds. Once the signal is triggered, the model evaluates the subsequent holding period performance, including average returns, win rates, and profit-loss ratios over 5-day and 10-day horizons[14][15] Model Evaluation: The model demonstrates strong performance in backtesting, with high win rates and favorable profit-loss ratios, indicating its effectiveness in capturing short-term rebounds[14][15] - Model Name: Platform Breakout Signal Model Construction Idea: This model identifies industries breaking out of consolidation phases, signaling potential upward momentum[14] Model Construction Process: The model monitors price movements to detect breakout patterns from established price ranges. Once a breakout is confirmed, the model tracks the performance over specific holding periods, analyzing metrics such as average returns, win rates, and profit-loss ratios over 5-day and 10-day horizons[14][15] Model Evaluation: The model shows moderate effectiveness, with reasonable win rates and profit-loss ratios, suggesting it can capture breakout opportunities but with less consistency compared to the oversold rebound signal[14][15] - Model Name: Stock Price Pressure Model Model Construction Idea: This model evaluates short-term selling pressure across industries to identify sectors with lower resistance to price increases[17] Model Construction Process: The model calculates theoretical selling pressure for each industry and ranks them accordingly. Industries with the lowest pressure are identified as having higher potential for price appreciation[17][18] Model Evaluation: The model effectively highlights sectors with lower selling pressure, which historically outperformed the market in backtesting[17][18] Model Backtesting Results - Oversold Rebound Signal - 5-day average return: 4.56% - 5-day win rate: 71% - 5-day profit-loss ratio: 1.87 - 10-day average return: 7.22% - 10-day win rate: 62% - 10-day profit-loss ratio: 1.66 - Average holding period: 5.2 days - Average holding return: 6.30% - Overall win rate: 76% - Overall profit-loss ratio: 2.3[14][15] - Platform Breakout Signal - 5-day average return: 3.30% - 5-day win rate: 62% - 5-day profit-loss ratio: 1.45 - 10-day average return: 5.10% - 10-day win rate: 61% - 10-day profit-loss ratio: 1.70 - Average holding period: 9 days - Average holding return: 6.77% - Overall win rate: 60% - Overall profit-loss ratio: 2.15[14][15] - Stock Price Pressure Model - Backtesting shows that holding the five industries with the lowest selling pressure generated approximately 15% excess returns relative to the Wind All A Index during the testing period[18][19]