形态学研究之十五：形态学在ETF轮动上的研究

Group 1 - The report discusses ETF rotation, which involves switching between different ETFs based on market conditions to adjust investment portfolios across various asset classes, industries, or regions [6][7][8] - The core logic of ETF rotation is to dynamically adjust the ETF portfolio to seek returns that exceed those of passive single-asset holdings [6][8] - The report introduces a methodology using stock patterns as raw data to synthesize ETF signals for rotation strategies, allowing for a comparative analysis of different ETFs [1][16] Group 2 - Three trading strategies based on historical data backtesting are proposed: fixed time point rebalancing, daily rebalancing, and a buy-and-hold exit strategy when signals disappear [2][35] - The buy-and-hold exit strategy is identified as the most effective, leveraging the characteristics of signal patterns while respecting low-frequency turnover rules [2][35] - The report provides performance metrics for each strategy, indicating that all strategies can outperform the Wind Mixed Equity Fund Index [2][35] Group 3 - The fixed time point rebalancing strategy shows varying annualized returns based on the number of ETFs selected, with the best performance observed when selecting six ETFs [17][20] - The daily rebalancing strategy also demonstrates performance metrics, with optimal results when selecting four ETFs, balancing risk and transaction costs [21][28] - The buy-and-hold exit strategy yields the highest annualized returns when selecting four ETFs, indicating that more is not always better in ETF rotation [29][31] Group 4 - The report emphasizes the importance of obtaining morphological signals through an API, allowing users to access comprehensive shape definitions and daily signals [36][37] - This API facilitates the replication of the index timing strategy, industry rotation strategy, and the ETF rotation strategy discussed in the report [36][37]