Model Of Limit Order Execution Probability And Order Placement Strategy

Nowadays,most equity and derivative exchanges are organized as orderdriven markets.In these markets,market participants trade against each other without the help of market makers or other intermediaries.Traders can either choose to execute their trade immediately at the best available prices by submitting market orders or to trade patiently by submitting limit orders.Undoubtedly,execution probability is one of the most important factors to consider between these two choices,as traders will have higher expected profit when execution probability increases.However,in the real markets,some orders are cancelled before they get executed.Discard these censored observations or treat them as unexecuted order will bias the execution probability estimation.The experiments in the simulation model indicate that survival analysis is the most appropriate method to accommodate censored observations.We conduct empirical analysis on Shenzhen Stocks Exchange with survival analysis for different groups of stocks.We create eight explanatory variables to capture the characteristics of the order book and find that the execution probability can be significantly influenced by these variables.This study also investigates the sensitivity between the execution probability and the explanatory variables.We find that the execution probability is very sensitive to the limit order price,but not sensitive to the limit order size.In the end,we utilize the model to predict execution probability.And by comparing to the static model,we find that traders can achieve better trading performance by incorporating execution probability prediction in their trading decisions.