Pairs trading: the case of Norwegian seafood companies

In this article, I investigate the performance of a pairs trading strategy on 18 seafood company stocks traded in the Norwegian consumer goods sector on the Oslo Stock Exchange. I apply both high-frequency and daily data from January 2005 to December 2014. I use two approaches – a distance approach and a cointegration approach – and compare the results. For both the distance and the cointegration approaches, nonconvergence of the pairs is high, which may indicate that more fundamental information about the companies traded should be accounted for. None of the strategies evaluated had significant profits after accounting for transaction costs. It therefore remains unclear which approach is best suited for pairs selection. Using high-frequency data yielded empirical distributions that were symmetrical and had a lower degree of leptokurtosis compared to the daily data.