An Analysis of Feasibility for Implementing Pairs Trading Strategy in Dhaka Stock Exchange using Vector Error Correction Model

Abstract

The objective of this study is to develop a financially profitable Pairs trading model for trading in Dhaka Stock Exchange. Pairs Trade is a statistical arbitrage investment strategy. The scope of this research is the primary stock market of Bangladesh – Dhaka Stock exchange. The study uses daily stock prices of a sample of 20 stocks listed in Dhaka Stock Exchange. The research first identifies a pair of stocks whose prices have a long-run equilibrium using Johansen’s test for Co-integration. The co-integrated stock pair is then modeled using a Vector Error Correction Model. The residual obtained from the estimated model serves as the guide to implementing Pairs Trade Strategy. The research finally identifies three pairs of stocks which have general long-run equilibriums. The residual series obtained from the Vector Error Correction Model for all three pairs are statistically significant. Based on this, the study implemented the pairs trade strategy using the residual series for a period of one to two months using real time data but doing hypothetical trading. It generated more than 100% (annual) rate of return for all three stock pairs. Finally, the study also compares the return using the Pairs Trade strategy with returns using the conventional financial analysis and found that pairs trading generated a much higher annualized returns compared to investment strategies using conventional financial analysis.