In this paper, we use the high frequency trading data of New York Stock Exchange to analyze the bid-ask spread components. It is found that there is an exponential relationship between the log returns of quoted midpoints and the trade volume. We also observe a negative linear correlation between the changes of quoted depth and the trade volume. Furthermore, changes of the quoted ask depth and the quoted bid depth are asymmetric due to the trading direction. Furthermore, statistical quality control charts, p-charts, are built for fixed number of trades to monitor unusual trades entering the market. Finally, logistic regression models are established to predict the probabilities of unusual trades entering the market based on the quotes and the quoted depth adjustments of the market makers.

In this study, we consider high frequency transaction data of NYSE, and apply statistical methods to characterize each trade into two classes, influential and ordinary liquidity trades. First, a median based approach is used to establish a high R-square price-volume model for high frequency data. Next, transactions are classified into four states based on the trade price, trade volume, quotes, and quoted depth. Volume weighted transition probability of the four states are investigated and shown to be distinct for informed trades and ordinary liquidity trades. Furthermore, four market reaction factors are introduced and studied. Logistic regression models of the influential trades are established based on the four factors and odds ratios are used to select the cutoff points.