Statistical Arbitrage: A Quantitative Trading Strategy

Share on Facebook0Tweet about this on Twitter0Share on LinkedIn17Share on Google+0

Statistical Arbitrage Explained

What is Quantitative Trading?

Quantitative trading is used to identify opportunities for trading by using statistical techniques and quantitative analysis of the historical data. Quantitative trading is applicable to information which is quantifiable like macroeconomic events and price data of securities.

Quantitative Trading models are used by Algo traders when trading of securities is based strictly on buy/sell decision of computer algorithms. An example of such a strategy which exploits quantitative techniques and is applied at Algorithmic trading desks is that of statistical arbitrage.

Statistical Arbitrage

StatArb or statistical arbitrage is a quantitative approach to equity trading involving data mining and statistical methods, as well as automated trading systems.

StatArb is actually any strategy that is bottom-up, beta/neutral in approach and uses statistical and econometric techniques in order to provide signals for execution. Signals are generated by a mean-reversion principle but can also be designed using such factors as lead/lag effects, corporate activity, short-term momentum, etc. This is referred as multi-factor approach to Statistical Arbitrage.

StatArb is an evolved version of pair-trading strategies, in which stocks are put into pairs by fundamental or market-based similarities. When one stock in a pair outperforms the other, the poorer performing stock is bought long with the expectation that it climbs its outperforming partner. The position is hedged from market changes/movements by shorting the other outperforming stock.  Because of the large number of stocks involved, the high portfolio turnover and the fairly small size of the spread one is trying to capture, the strategy is often implemented in an automated fashion and great attention is placed on reducing trading costs. Statistical arbitrage has become a major force at both hedge funds and investment banks.

Implementation steps of a statistical arbitrage strategy

Figure 1: Implementation steps of a statistical arbitrage strategy

How Statistical Arbitrage Works?

Securities such as stocks tend to trade in upward and downward cycles and a quantitative method seeks to capitalize on those trends. Trending behavior of quantitative trading uses software programs to track patterns or trends. Trends uncovered are based on the volume and the frequency the price of a security at which it is traded.

Statistical Arbitrage between two stocks under Cement Industry

Figure 2: Statistical Arbitrage between two stocks under “Cement” Industry: ACC and Ambuja both listed at National Stock Exchange of India.

In the image above, the stock prices of ACC and Ambuja are represented over a period of six years. You can see both the stocks stay quite close to each other during the entire time span, with only a few certain instances of separation. It is in those separation periods that an arbitrage opportunity arises based on an assumption that the stock prices with move closer again.

The crux in identifying such opportunities lies in two main factors:

  • Identifying the pairs which require advanced time series analysis and statistical tests
  • Specifying the entry-exit points for the strategy to leverage on the market position

There are plenty of in-built pair trading indicators on popular platforms to identify and trade in pairs. However, many a times, transaction cost which is a crucial factor in earning profits from a strategy, is usually not taken in account in calculating the projected returns. Therefore, it is recommended that traders make their own statistical arbitrage strategies keeping into account all the factors at the time of backtesting which will affect the final profitability of the trade.

Next Step

Start learning more quantitative strategies from Basic Statistics for Trading Strategies (Part 1). You may also start by understanding ‘market microstructure’ and ‘risk management in automated trading’.

Share on Facebook0Tweet about this on Twitter0Share on LinkedIn17Share on Google+0

Leave a Reply

Your email address will not be published. Required fields are marked *