Online Learning and Statistical Tests for Technical Trading Strategies

Summary

This study uses adversarial expert-based online learning to select parameters for zero-cost portfolio strategies and to combine technical trading strategies into an aggregate portfolio. It examines strategy populations on daily and intraday Johannesburg Stock Exchange data, then uses unsupervised learning to reduce and visualize their changing composition. The analysis also accounts for trading costs and slippage, compares performance with an online benchmark portfolio algorithm, and estimates the risk of back-test overfitting.

The aggregate strategies are assessed for statistical arbitrage using a hypothesis test, with different outcomes across sampling frequencies: daily strategies fail the tests after costs, while intraday strategies are not falsified as statistical arbitrages after costs. That distinction is evidence from the historical datasets and procedures used in the study, not proof of future profitability. The reported conclusion depends on the chosen strategies, market data, cost and slippage estimates, and statistical tests; the excerpt does not provide detailed numerical performance results.

Key ideas

• An online learning algorithm selects parameters and combines technical trading strategies into a portfolio.
• The study examines daily and intraday Johannesburg Stock Exchange data.
• Daily aggregate strategies fail statistical arbitrage tests after costs, while intraday strategies are not falsified.
• The analysis considers slippage, transaction costs, a benchmark algorithm, and back-test overfitting.
• Historical statistical results do not establish that the strategies will remain profitable.

Tags

# Learning the dynamics of technical trading strategies


# Learning the dynamics of technical trading strategies









We use an adversarial expert based online learning algorithm to learn the optimal parameters required to maximise wealth trading zero-cost portfolio strategies. The learning algorithm is used to determine the relative population dynamics of technical trading strategies that can survive historical back-testing as well as form an overall aggregated portfolio trading strategy from the set of underlying trading strategies implemented on daily and intraday Johannesburg Stock Exchange data. The resulting population time-series are investigated using unsupervised learning for dimensionality reduction and visualisation. A key contribution is that the overall aggregated trading strategies are tested for statistical arbitrage using a novel hypothesis test proposed by Jarrow et al. (2012) on both daily sampled and intraday time-scales. The (low frequency) daily sampled strategies fail the arbitrage tests after costs, while the (high frequency) intraday sampled strategies are not falsified as statistical arbitrages after costs. The estimates of trading strategy success, cost of trading and slippage are considered along with an online benchmark portfolio algorithm for performance comparison. In addition, the algorithms generalisation error is analysed by recovering a probability of back-test overfitting estimate using a nonparametric procedure introduced by Bailey et al. (2016). The work aims to explore and better understand the interplay between different technical trading strategies from a data-informed perspective.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.