Title: Learning zero-cost portfolio selection with pattern matching for intraday trading
Speaker: Tim Gebbie
Authors: Tim Gebbie, Fayyaaz Loonat
Abstract: We consider and extend an adversarial agent-based learning approach to the situation of zero-cost portfolio selection in the domain of quantitative trading. The algorithms are directly compared to standard NYSE test cases from prior literature. The learning algorithm is used to select parameters for agents (or experts) generated by pattern matching past dynamics using a simple nearest-neighbour search algorithm. We demonstrate that patterns in financial time-series on the JSE can be systematically exploited in collective, but that this does not imply predictability of the individual asset time-series themselves. We show that these types of machine learning algorithms are well suited for intra-day quantitative trading.