行业轮动周报：双创涨速明显提升，ETF资金配置思路偏补涨-20250901

• Model Name: Diffusion Index Model; Construction Idea: The model is based on the principle of price momentum, capturing industry trends through diffusion indices; Construction Process: The model tracks the weekly and monthly performance of various industries, calculating the diffusion index for each industry. The formula for the diffusion index is not explicitly provided in the report; Evaluation: The model has shown varying performance over the years, with significant drawdowns during market reversals[25][26][29] - Model Name: GRU Factor Model; Construction Idea: The model leverages GRU (Gated Recurrent Unit) deep learning networks to process minute-level volume and price data, aiming to capture trading information; Construction Process: The model ranks industries based on GRU factors, which are derived from the deep learning network's analysis of trading data. The specific formula for GRU factors is not provided in the report; Evaluation: The model has struggled to capture excess returns in a focused market environment, particularly in 2025[32][33][37] Model Backtest Results - Diffusion Index Model, Average Weekly Return: 2.97%, Excess Return over Equal-Weighted Index: 1.94%, August Excess Return: 4.54%, Year-to-Date Excess Return: 5.08%[29] - GRU Factor Model, Average Weekly Return: 1.85%, Excess Return over Equal-Weighted Index: 0.93%, August Excess Return: -2.53%, Year-to-Date Excess Return: -7.65%[37] Factor Construction and Evaluation - Factor Name: Diffusion Index; Construction Idea: The factor is constructed based on the momentum of industry prices, capturing the upward or downward trends; Construction Process: The diffusion index is calculated weekly and monthly for each industry, ranking them accordingly. The specific calculation method is not detailed in the report; Evaluation: The factor has shown mixed performance, with significant drawdowns during market reversals[25][26][29] - Factor Name: GRU Factor; Construction Idea: The factor is derived from GRU deep learning networks analyzing minute-level trading data; Construction Process: The GRU factor ranks industries based on the network's analysis, with higher ranks indicating stronger trading signals. The specific calculation method is not detailed in the report; Evaluation: The factor has struggled to capture excess returns in a focused market environment, particularly in 2025[32][33][37] Factor Backtest Results - Diffusion Index Factor, Top Industries: Comprehensive (1.0), Nonferrous Metals (0.973), Communication (0.971), Banking (0.965), Media (0.945), Retail (0.916)[26] - GRU Factor, Top Industries: Petroleum and Petrochemical (3.38), Non-Banking Financial (3.16), Retail (2.59), Food and Beverage (1.29), Electric Power and Utilities (0.21), Coal (0.16)[33]