ETF Arbitrage And Return Predictability In China

Demand shocks cause assets to deviate from their intrinsic value,making identical securities have different pricing,which triggers investors’ arbitrage behavior.From a large amount of historical data,it is not difficult to see that this arbitrage behavior of investors can make asset prices back to intrinsic value to some extent,but it is not completely equal to intrinsic value,that is,this return of value is volatile,gradual,and asymptotic.After arbitrage behaviors,market information cannot be immediately reflected in prices.This phenomenon triggered the author’s thinking.Since market information cannot be immediately reflected in prices,can arbitrage behavior,a leading indicator,be used to predict the future trend of asset returns?The author uses the 20052019 ETF and its constituent stocks as the research obj ect to study whether the arbitrage behavior of the ETF market has a predictive effect on the return rate of ETF itself and its constituent stocks.First,this paper constructs a monthly indicator ETFarb to measure the changes in shares outstanding caused by arbitrage,then uses this indicator to group ETFs to study whether the next month’s returns between different groups have a significant difference.The conclusion is that ETF arbitrage has a negative predictive effect for ETF return.Based on one-dimensional grouping result above,the author adds indicators such as market capitalization,transaction volume,and last month’s return to perform twodimensional grouping.The grouping results show that some indicators have significant effects on ETF return.Finally,this paper uses Fama-Mecbeth regression for further test.Similarly,when studying the predictability of ETF arbitrage on its constituent stock returns,this paper constructs the indicator StockArb,and the results show that ETF arbitrage also has a strong negative predictive effect on the returns of individual stocks.