Measuring Financial Time Series Similarity With a View to Identifying Profitable Stock Market Opportunities

TL;DR

This paper introduces a novel similarity metric for historical stock price data, enhancing case-based prediction methods to better identify profitable trading opportunities in volatile markets.

Contribution

It develops a new similarity metric tailored for stock time series, improving case-based forecasting accuracy over traditional methods.

Findings

The new similarity metric outperforms Euclidean and correlation-based measures.

Case-based approach with the new metric yields better prediction results.

Demonstrated effectiveness in real-world stock market data.

Abstract

Forecasting stock returns is a challenging problem due to the highly stochastic nature of the market and the vast array of factors and events that can influence trading volume and prices. Nevertheless it has proven to be an attractive target for machine learning research because of the potential for even modest levels of prediction accuracy to deliver significant benefits. In this paper, we describe a case-based reasoning approach to predicting stock market returns using only historical pricing data. We argue that one of the impediments for case-based stock prediction has been the lack of a suitable similarity metric when it comes to identifying similar pricing histories as the basis for a future prediction -- traditional Euclidean and correlation based approaches are not effective for a variety of reasons -- and in this regard, a key contribution of this work is the development of a novel similarity metric for comparing historical pricing data. We demonstrate the benefits of this metric and the case-based approach in a real-world application in comparison to a variety of conventional benchmarks.