• CHETAN TAYAL
• Lalitha V.P

Pairs Trading is employed widely as a market neutral trading strategy using the concepts of statistical arbitrage. The theory behind this technique is based on the idea of mean-reverting time series and is designed to illustrate the relationship between the two assets. In order to minimise risk when the market goes in only one direction, investors may benefit by profiting on one side of the bet. One of the hardest aspects of pairs trading is deciding which pairs have a link that can be called fundamental and which are merely coincidental. The proposed machine learning method in this research is unsupervised clustering for construction of the search space for pair selection, and an algorithm to evaluate the search space is presented. In an effort to recognise profitable trading pairs, not only do we realise we are picking out better pairs, these pairs have less exposure to the market and are less volatile, thus presenting less risk to an investor.

Pairs Trading, OPTICS, Unsupervised Clustering, Spread, Backtest

"Application of Machine Learning in Market-Neutral Pairs Trading in the Indian Stock Exchange", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.8, Issue 7, page no.c430-c439, July-2021, Available :http://www.jetir.org/papers/JETIR2107324.pdf

"Application of Machine Learning in Market-Neutral Pairs Trading in the Indian Stock Exchange", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.8, Issue 7, page no. ppc430-c439, July-2021, Available at : http://www.jetir.org/papers/JETIR2107324.pdf