An order flow model and a liquidity measure of financial markets

The thesis seeks a better understanding of liquidity generation process of financial markets and attempts to find a quantitative measure of market liquidity. Various statistical modeling techniques are introduced to model order flow generation, which is a liquidity generation process of the market. The order flow model successively replicates various statistical properties of price returns including fat-tailed distribution of returns, no autocorrelation of returns and strong positive autocorrelation of transaction signs. While attempting to explain how the order flow model satisfies Efficient Market Hypothesis (EMH), I discovered a method of calibrating market liquidity from order flow data. Thesis Supervisor: Tomaso Poggio Title: Eugene McDermott Professor Thesis Supervisor: Andrew W. Lo Title: Harris and Harris Group Professor Thesis Supervisor: J. Doyne Farmer Title: Professor, Santa Fe Institute