Search

We consider strategies which use a collection of popular technical indicators as input and seek a profitable trading rule defined in terms of them. We consider two popular computational learning approaches, reinforcement learning and genetic programming, and compare them to a pair of simpler methods: the exact solution of an appropriate Markov decision problem, and a simple heuristic. We find that although all methods are able to generate significant in-sample and out-of-sample profits when transaction costs are zero, the genetic algorithm approach is superior for non-zero transaction costs, although none of the methods produce significant profits at realistic transaction costs. We also find that there is a substantial danger of overfitting if in-sample learning is not constrained.

This paper introduces the use of dynamic stochastic optimisation for pension fund management. The design of such products involves econometric modelling, economic scenario generation, generic methods of solving optimization problems and modelling of required risk tolerances. In nearly all the historical backtests using data over roughly the past decade the system described (with transactions costs taken into account) outperformed the benchmark S&P500.