Utilizing Price Convergence in Day Trading Strategies for Smaller Stocks on the JSE

Day traders often find it useful to trade smaller stocks in anticipation of them mimicking larger stocks’ price movements in the same sector. In order to do so successfully, the trader will have to know how long it takes for the price of the smaller stock to converge to that of the dominant stock in that sector. Permutations of the GARCH family are often used to describe such price and volatility spillover effects between stocks. This approach is very useful when testing for such effects on returns data, but less so when trying to measure the speed of adjustment between two stocks on non-differenced data. When utilising GARCH models, it is necessary to work with stationary (and thus often differenced returns) data. However, when trading, day traders ignore the movement of returns data and focus on the price movements as quoted on the bourse. Since most stock price time series contain a unit root in their non-differenced (level) state, this study uses Johansen’s cointegration test to measure the speed of adjustment between a large stock (expressed in size of capitalization) and that of a small stock on the JSE Securities Exchange South Africa. The results of these tests are used to advise day traders on the price convergence of such stocks in order to profit from them.