A hybrid learning approach to detecting regime switches in financial markets

Abstract

Financial markets are of much interest to researchers due to their dynamic and stochastic nature. With their relations to world populations, global economies and asset valuations, understanding, identifying and forecasting trends and regimes are highly important. Attempts have been made to forecast market trends by employing machine learning methodologies, while statistical techniques have been the primary methods used in developing market regime switching models used for trading and hedging. In this paper we present a novel framework for the detection of regime switches within the US financial markets. Principal component analysis is applied for dimensionality reduction and the 𝑘-means algorithm is used as a clustering technique. Using a combination of cluster analysis and classification, we identify regimes in financial markets based on publicly available economic data. We display the efficacy of the framework by constructing and assessing the performance of two trading strategies based on detected regimes. CCS CONCEPTS• Information systems → Clustering.

HireVAE: An Online and Adaptive Factor Model Based on Hierarchical and Regime-Switch VAE

“…Machine learning based methods open another main direction in modeling regime switching, such as Hidden Markov Model [Wang et al, 2020] and Gaussian Mixed Models [Botte and Bao, 2021]. As a hybrid method, [Akioyamen et al, 2020] applies principal component analysis and k-means clustering to identify regimes in financial markets. Deep learning based regime switching models of energy commodity prices [Mari and Mari, 2022].…”

“…Both practitioners and academics make many attempts to model the regime change in markets using statistical learning [Guidolin and Timmermann, 2007], machine learning [Uysal and Mulvey, 2021], or hybrid methods [Akioyamen et al, 2020] from different perspectives of the market. Existing data-driven approaches view regime switching as a clustering problem and represent each regime as a cluster, subdividing the market regime based on the information absorbed from the entire training data.…”