Quantitative Models and Construction Methods 1. Model Name: FuGuo ETF Rotation Strategy - **Model Construction Idea**: The strategy is based on the FuGuo ETF rotation factor, which evaluates ETFs' investment value from four dimensions: profitability, operational quality, valuation momentum, and analyst expectations. These dimensions are combined into a composite rotation factor through standardization and equal weighting[26][30]. - **Model Construction Process**: 1. **Profitability Factors**: - **Excluding Non-recurring Profit Growth (QoQ)**: Measures the quarterly change in net profit after excluding non-recurring items, aggregated using the median method[30][31]. - **Net Profit Growth (YoY)**: Measures the year-over-year change in net profit, aggregated using the median method[30][31]. 2. **Operational Quality Factors**: - **Operating Capital Turnover**: Ratio of operating capital to operating revenue, calculated semi-annually using the "leading stock" method[30][31]. - **Operating Capital Proportion**: Ratio of operating capital to total assets, calculated year-over-year using the "leading stock" method[30][31]. 3. **Valuation Momentum Factor**: - **Inverse Price-to-Earnings Ratio**: Measures the semi-annual change in the inverse P/E ratio, reflecting market sentiment[30][31]. 4. **Analyst Expectation Factor**: - **Analyst Forecast Change**: Tracks the 3-month change in analysts' consensus EPS forecasts, aggregated using the "leading stock" method[30][31]. 5. The above six factors are standardized and equally weighted to form the FuGuo ETF rotation factor[26][30]. - **Model Evaluation**: The FuGuo ETF rotation factor demonstrates strong predictive power for ETF performance, with stable IC values and effective multi-dimensional evaluation of ETF investment value[26][30]. --- Model Backtesting Results 1. FuGuo ETF Rotation Strategy - **Annualized Return**: 7.26%[19] - **Annualized Volatility**: 21.92%[19] - **Sharpe Ratio**: 0.33[19] - **Maximum Drawdown**: 42.20%[19] - **Turnover Rate (Monthly)**: 53.21%[19] - **Annualized Excess Return**: 5.49%[19] - **Tracking Error**: 9.27%[19] - **Information Ratio (IR)**: 0.59[19] - **Excess Maximum Drawdown**: 15.23%[19] - **May 2025 Return**: -2.03%[19] - **May 2025 Excess Return**: -3.46%[19] --- Quantitative Factors and Construction Methods 1. Factor Name: FuGuo ETF Rotation Factor - **Factor Construction Idea**: The factor evaluates ETFs' investment value by integrating profitability, operational quality, valuation momentum, and analyst expectations into a composite score[26][30]. - **Factor Construction Process**: 1. **Profitability Factors**: - **Excluding Non-recurring Profit Growth (QoQ)**: Measures the quarterly change in net profit after excluding non-recurring items, aggregated using the median method[30][31]. - **Net Profit Growth (YoY)**: Measures the year-over-year change in net profit, aggregated using the median method[30][31]. 2. **Operational Quality Factors**: - **Operating Capital Turnover**: Ratio of operating capital to operating revenue, calculated semi-annually using the "leading stock" method[30][31]. - **Operating Capital Proportion**: Ratio of operating capital to total assets, calculated year-over-year using the "leading stock" method[30][31]. 3. **Valuation Momentum Factor**: - **Inverse Price-to-Earnings Ratio**: Measures the semi-annual change in the inverse P/E ratio, reflecting market sentiment[30][31]. 4. **Analyst Expectation Factor**: - **Analyst Forecast Change**: Tracks the 3-month change in analysts' consensus EPS forecasts, aggregated using the "leading stock" method[30][31]. 5. The above six factors are standardized and equally weighted to form the FuGuo ETF rotation factor[26][30]. - **Factor Evaluation**: The factor demonstrates stable performance with an average IC of 6.80% and a risk-adjusted IC of 0.22, indicating its effectiveness in predicting ETF performance[12][14]. --- Factor Backtesting Results 1. FuGuo ETF Rotation Factor - **Average IC**: 6.80%[12] - **Standard Deviation of IC**: 31.51%[12] - **Minimum IC**: -59.85%[12] - **Maximum IC**: 78.19%[12] - **Risk-adjusted IC**: 0.22[12] - **T-statistic**: 2.24[12] - **May 2025 IC**: -30.91%[14] - **May 2025 Long-Short Portfolio Return**: -6.96%[14]

证 券 研 究 报 告 【点评报告】 形态学研究之十五：形态学在 ETF 轮动上的研究 ❖ 摘要 ETF 轮动，是指在不同 ETF 之间切换，根据市场情况调整投资组合。可涉及 到不同资产类别、行业或地区。比如如何选择 ETF，轮动的依据是什么，比如 动量、估值、宏观经济指标等。 本篇，我们提出了使用形态作为原始数据，将 ETF 成分股合成 ETF 信号的方 式来做 ETF 轮动。在先前报告中，我们已经成功基于成分股的形态构建了指 数、ETF 的择时策略，现在就是设计指标将同一截面下的各个 ETF 指数数据 可比，就可以构建 ETF 轮动策略。 当我们把 ETF 指数每天的多空个股个数求和后，除以当时该 ETF 指数的成分 股个数，在进行 30 天的 HMA 均线求解后，该指标就表征了该 ETF 从结构上 形态学的看多或者看空力量的变化与对比，由于都除以了成分股个数，也消除 了不同 ETF 因为成分股个数不同不可比的情况。 基于历史数据回测，不管是固定时间点调仓、每日调仓、或者是买入持有信号 消失退出策略等，都可跑赢万得偏股混合型基金指数。固定时间点调仓策略优 势在于计算量少，交易次数低，手续费少，但收益率也一 ...