Title: Estimation of market impact cost using high frequency execution and order book data
Authors: Kenta Yamada - National Institute of Informatics (Japan) [presenting]
Takayuki Mizuno - National Institute of Informatics (Japan)
Abstract: The aim is to analyze the historical data set of the Tokyo Stock Exchange(TSE) for a 29-month period from August 2014 to December 2016 which includes every transaction and order book snapshot, and found two major relationships: (i) a proportional relationship between the return of the market price and the order imbalance between buying and selling market orders, where this relationship describes market impact, and (ii) an inverse proportional relationship between the market impact and the averaged volumes of limit orders on the order book. In this analysis, we focus on daily and monthly time scale, and we found the market impact depends on the stock and period, however applying the binomial test to the size of the market impact among stocks, it is statistically shown that market impact depends on the stocks. In the analysis of (ii), we show the market impact depends on the order book volume. The results of our analysis imply that when we estimate transaction costs, we need to consider not only a commission fee and spread cost but also a market impact cost, and therefore we also introduce estimation results of transaction costs for a practical application of our studies.