金融工程周报：主力资金流入汽车行业，情绪高涨-20250509

Quantitative Models and Construction Methods 1. Model Name: A-Share Industry Rotation Model - Model Construction Idea: The model is built using six factors: capital, valuation, sentiment, momentum, overbought/oversold, and profitability, to create a scoring system for evaluating industries comprehensively [13] - Model Construction Process: - Capital Factor: Uses the main net inflow rate of industry funds as the primary data source - Valuation Factor: Based on the valuation percentile of the industry over the past year - Sentiment Factor: Proportion of rising constituent stocks serves as the main data source - Momentum Factor: Uses the MACD indicator as the primary data source - Overbought/Oversold Factor: Relies on the RSI indicator as the key data source - Profitability Factor: Based on the consensus forecast EPS percentile of the industry over the past year [13] - Model Evaluation: The model provides a comprehensive scoring mechanism to assess industry performance, effectively identifying high-performing and low-performing industries [13] 2. Model Name: Consensus Stock Selection Model - Model Construction Idea: The model combines momentum, valuation, and price increase frequency factors with high-frequency capital flow data to select stocks with the highest similarity between capital flow trends and stock price trends [16] - Model Construction Process: - Identify high-performing secondary industries over the past 30 days - Calculate momentum factors, valuation factors, and price increase frequency for stocks within these industries - Use high-frequency minute-level data to calculate capital inflow/outflow changes for each stock - Select the top three secondary industries and pick five stocks from each based on the highest similarity between capital flow trends and stock price trends [16] --- Model Backtesting Results 1. A-Share Industry Rotation Model - Top Scoring Industries: Non-bank financials, communication, and automotive scored the highest, with total scores of 8, 8, and 7, respectively - Low Scoring Industries: Building materials, social services, and steel scored the lowest, with total scores of -10, -5, and -5, respectively [14][15] 2. Consensus Stock Selection Model - Selected Industries: Feed, precious metals, and animal health II were identified as the top-performing secondary industries - Selected Stocks: - Feed: Baiyang Co., Zhongchong Co., Zhenghong Technology, Bangji Technology, Guibao Pet - Precious Metals: Hunan Silver, Xiaocheng Technology, Western Gold, Hunan Gold, Shandong Gold - Animal Health II: Jinhe Bio, Haili Bio, Kexian Bio, Shengwu Co., Ruipu Bio [17] --- Quantitative Factors and Construction Methods 1. Factor Name: Capital Factor - Construction Idea: Measures the main net inflow rate of industry funds to assess capital movement trends [13] - Construction Process: Calculate the net inflow of main funds for each industry and normalize the data to create a scoring metric [13] 2. Factor Name: Valuation Factor - Construction Idea: Evaluates the relative valuation of an industry based on its historical percentile over the past year [13] - Construction Process: Calculate the percentile of the industry's valuation over the past year and assign scores accordingly [13] 3. Factor Name: Sentiment Factor - Construction Idea: Uses the proportion of rising constituent stocks to gauge market sentiment [13] - Construction Process: Calculate the percentage of rising stocks within an industry and normalize the data for scoring [13] 4. Factor Name: Momentum Factor - Construction Idea: Leverages the MACD indicator to measure momentum trends [13] - Construction Process: Compute the MACD values for each industry and assign scores based on the strength of momentum [13] 5. Factor Name: Overbought/Oversold Factor - Construction Idea: Uses the RSI indicator to identify overbought or oversold conditions [13] - Construction Process: Calculate RSI values for each industry and assign scores based on thresholds for overbought/oversold conditions [13] 6. Factor Name: Profitability Factor - Construction Idea: Assesses profitability using the consensus forecast EPS percentile over the past year [13] - Construction Process: Calculate the percentile of the consensus forecast EPS for each industry and assign scores accordingly [13] --- Factor Backtesting Results 1. Capital Factor - Top Scoring Industries: Automotive (+3), Non-bank financials (+2), Communication (+2) [15] 2. Valuation Factor - Top Scoring Industries: Automotive (+3), Non-bank financials (+3), Communication (+2) [15] 3. Sentiment Factor - Top Scoring Industries: Communication (+3), Automotive (+1), Non-bank financials (-2) [15] 4. Momentum Factor - Top Scoring Industries: Non-bank financials (+3), Communication (+1), Automotive (+1) [15] 5. Overbought/Oversold Factor - Top Scoring Industries: Non-bank financials (+3), Communication (+3), Automotive (+3) [15] 6. Profitability Factor - Top Scoring Industries: Automotive (+3), Non-bank financials (+3), Communication (+3) [15]