A mixed-frequency VAR application to studying joint dynamics of foreign investor trading and stock market returns

Abstract

We present the first application of the mixed-frequency VAR (MF-VAR) method in the market microstructure literature, studying the interaction between stock market returns and foreign investors' trading. MF-VAR allows us to use daily investor trading data together with higher-frequency return series and uncover novel intraday patterns in the feedback trading behavior and the information content of trading. Using data from Korea, we find that foreign investors chase opening-hour returns, and their trading has the ability to forecast subsequent days' late-hour returns. This pattern suggests that foreign investors selectively respond to the information incorporated during opening hours. Over the years, foreign investors' response to intraday returns has become more prompt, and the predictive ability of their trading has disappeared. A specific test made feasible by the MF-VAR method does not support the global private information hypothesis.