Optimisation of Trading Strategies using Parameterised Decision Rules

In the context of a dynamic trading strategy, the ultimate purpose of any forecasting model is to choose actions which result in the optimisation of the trading objective. In this paper we develop a methodology for optimising an objective function, using a parameterised decision rule, for a given forecasting model. We simulate the expected trading performance for different decision parameters and levels of prediction accuracy. We then apply the technique to a forecasting model of mispricing within a group of equity indices. We show that optimisation of the proposed decision rule can increase the annualised Sharpe Ratio by a factor of 1.7 over a naive decision rule.