This study uses the Johansen test for cointegration to select trading pairs for use within a pairs trading framework. A long-run equilibrium price relationship is then estimated for the identified trading pairs, and the resulting mean-reverting residual spread is modeled as a Vector-Error-Correction model (VECM). The study uses 5 years of daily stock prices starting from the beginning of July, 2002. The search for trading pairs is restricted to 17 financial stocks listed on the ASX200. The results show that two cointegrated stocks can be combined in a certain linear combination so that the dynamics of the resulting portfolio are governed by a stationary process. Although a trading rule is not employed to access the profitability of this trading strategy, plots of the residual series
show a high rate of zero crossings and large deviations around the mean. This would
suggest that this strategy would likely be profitable. It can also be concluded that in the presence of cointegration, at least one of the speed of adjustment coefficients must be significantly different from zero.