Dynamics of the price-volume information flow based on surrogate time series.

This paper uses transfer entropy and surrogates to analyze the information flow between price and transaction volume. We use random surrogates to construct local random permutation (LRP) surrogates that can analyze the local information flow in detail. The analysis based on the toy models verifies the effectiveness of the LRP method. We further apply it to analyze three financial datasets, including two index datasets and one stock dataset. Empirical analysis shows that both the S&P500 index data and SSEC index data include rich information flow dynamics. There was a stronger information flow during the stock bubble burst or the financial crisis. In addition, tests based on stock data suggest that market crises may lead to changes in the relationship between prices and trading volume. This paper provides a new way to analyze the price-volume relationship, which can effectively detect the drastic changes in the local information flow, thereby providing a method for studying the impact of events.