Open AccessWhich Pairs of Stocks should we Trade? Selection of Pairs for Statistical Arbitrage and Pairs Trading in Karachi Stock Exchange
Author(s) -
Laila Taskeen Qazi,
Atta Ur Rahman,
Saleem Gul
Publication year - 2015
Publication title -
pakistan development review
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.154
H-Index - 26
ISSN - 0030-9729
DOI - 10.30541/v54i3pp.215-244
Subject(s) - statistical arbitrage , index arbitrage , pairs trade , cointegration , econometrics , economics , arbitrage , residual , stock exchange , granger causality , financial economics , stock (firearms) , mean reversion , mathematics , algorithmic trading , finance , risk arbitrage , arbitrage pricing theory , capital asset pricing model , mechanical engineering , alternative trading system , algorithm , engineering
Pairs Trading refers to a statistical arbitrage approachdevised to take advantage from short term fluctuations simultaneouslydepicted by two stocks from long run equilibrium position. In this studya technique has been designed for the selection of pairs for pairstrading strategy. Engle-Granger 2-step Cointegration approach has beenapplied for identifying the trading pairs. The data employed in thisstudy comprised of daily stock prices of Commercial Banks and FinancialServices Sector. Restricted pairs have been formed out of highly liquidlog share price series of 22 Commercial Banks and 19 Financial Servicescompanies listed on Karachi Stock Exchange. Sample time period extendedfrom November 2, 2009 to June 28, 2013 having total 911 observations foreach share prices series incorporated in the study. Out of 231 pairs ofcommercial banks 25 were found cointegrated whereas 40 cointegratedpairs were identified among 156 pairs formed in Financial ServicesSector. Furthermore a Cointegration relationship was estimated byregressing one stock price series on another, whereas the order ofregression is accessed through Granger Causality Test. The meanreverting residual of Cointegration regression is modeled through theVector Error Correction Model in order to assess the speed of adjustmentcoefficient for the statistical arbitrage opportunity. The findings ofthe study depict that the cointegrated stocks can be combined linearlyin a long/short portfolio having stationary dynamics. Although for thegiven strategy profitability has not been assessed in this study yet theVECM results for residual series show significant deviations around themean which identify the statistical arbitrage opportunity and ensureprofitability of the pairs trading strategy. JEL classifications: C32,C53, G17 Keywords: Pairs Trading, Statistical Arbitrage, Engle-Granger2-step Cointegration Approach, VECM.