Seminar by Dr. Binh Do
- Topic: Cointegration, Statistical Arbitrage and Weak-Form Market Efficiency
- Speaker: Dr. Binh Do, Senior Lecturer, Department of Accounting and Finance, Monash University
- When: 5th October, 2012, (Fri)
Statistical arbitrage is a zero-cost trading strategy that offers positive expected payoffs whilst still allowing for negative outcomes. Amongst the hedge fund circle, it refers to a class of strategies that seek to exploit short-term mispricings between two or more assets. In this paper, I investigate the empirical performance of statistical arbitrage that uses cointegration to model co-movements amongst non-stationary prices. Using the U.S. daily stock data over period 1962-2011, I find that the cointegration-based strategies generate positive and statistically significant excess returns, however their profitability has materially declined since 2000. This profit decline coincides with a drop in the frequency of cointegrations observed for all formation groups considered: 2, 3 and 4 stocks. The evidence suggests that the market is either increasingly more efficient or more risky for this type of trading. I also find that, under this cointegration approach, basket trading amongst 3 and 4 stocks experiences lower volatility than pairs trading.
View the full paper: Cointegration, Statistical Arbitrage and Weak-Form Market Efficiency (PDF, 389.3KB)