The paper applies machine learning tools in pairs trading. Three different algorithms, namely, Support Vector Machine (SVM), Random Forest (RF) and Adaptive Neuro Fuzzy Inference System (ANFIS), have been used for predictive modeling of the value of the ratio of share prices of pairs of companies. The study considers nine different independent variables/features for forecasting. The analytical framework combines the mean reverting property of the movement of a pair of prices along with technical indicators. We also use feature selection algorithms for justification of the nine independent variables. The results support our methodology and also selection of the features for prediction.

Pairs Trading is an arbitrage strategy which generates profits from the movement in the ratio of prices of two stocks. Ideally, the ratio of the price of the two stocks, which belong to the same sector and whose prices are highly correlated, should be stable. However, given the inherent randomness of stock price movements, this ratio may fluctuate. The fundamental basis of pairs trading is that although there would be fluctuations in the ratio of the prices, it would be mean reverting. Thus if the ratio rises, it is expected that it would fall and if the ratio falls, then it is expected to rise. For the ratio to rise, one reason could be that the price of one stock is rising at a faster rate than the other one, in spite of the fact that they are from the same sector and are having highly correlated prices. The trading strategy in this case is to short the faster moving stock and long the slower moving stock; the expectation being that the stock whose price is rising faster will fall, and the stock whose price is not rising that fast, will continue to rise. Similarly, if the ratio falls, then we long the faster falling one and short the slower falling one. Clearly, if a framework could be devised to predict the future movement in the ratio of stock prices, it would be helpful for such trading.

Applied Finance Journal, Application of Machine Learning Tools, Predictive Modeling of Pairs Trade, Indian Stock Market