An MA-MRR model for transaction-level analysis of highfrequency trading processes

The transaction-level analysis of security price changes by Madhavan,Richardson,and Roomans(1997,hereafter MRR)is a useful framework for financial analysis.The first-order Markov property of trading indicator variables is a critical assumption in the MRR model,which contradicts the information lag empirically demonstrated in high-frequency trading processes.In this study,a nonparametric test is employed,which shows that the Markov property of the trading indicator variables is rejected on most trading days.Based on the spread decomposed structure,an MA-MRR model was proposed with a moving average structure adopted to absorb the information lag as an extension.The empirical results show that the information lag plays an important role in measuring the adverse selection risk parameter and that the difference in this parameter between the original and the extension is significant.Furthermore,our analysis suggests that the information lag parameter is a useful measure of the average speed at which information is incorporated into prices.