Applications of Copulas and Machine Learning Algorithms on Pairs Trading
DOI:
https://doi.org/10.54691/bcpbm.v46i.5070Keywords:
Pairs Trading, Bollinger Bands, Copulas, Support Vector Machines, RUT, SP400.Abstract
This paper investigated three different pairs trading strategies: the usual baseline (linear) approach, the copulas method, and the machine learning technique. We selected two equity indexes, Russell 2000 (RUT) and S&P400 (SP400), for the pairs trading. It is found that the most significant reasons financial companies employ pairs trading are either its stable nature or profitability. In addition, during a recession session (eg. the Covid-19), the cumulative return of pairs trading can outperform the usual buy-hold strategy and even the market smoothly.
Downloads
References
Appel, Gerard (2008). Understanding Macd (Moving Average Convergence Divergence).
Bilgili, Faik (1998). Stationarity and cointegration tests: Comparison of Engle - Granger and Johansen methodologies.
Géron, Aurélien (2019). Hands-on machine learning with Scikit-Learn, Keras and TensorFlow: concepts, tools, and techniques to build intelligent systems. 2nd ed.
J. W. Wilder (1978). New Concepts in Technical Trading Systems.
Liew, R., & Wu, Y (2013). Pairs trading: A copula approach. J Deriv Hedge Funds 19, 12–30 (2013).
Ling, Hu (2006). Dependence patterns across financial markets: a mixed copula approach.
Mitchell, Cory (2022). Using Bollinger Bands to Gauge Trends.
Montana, G., & Parella, F. (2009). Data Mining for Algorithmic Asset Management.
Radovanovic, Igor (2022). Cluster Analysis – Machine Learning for Pairs Trading.
Szklarz, J., Rosillo, R., Alvarez, N., Fernández, I., Garcia, N. (2018). Application of Support Vector Machine on Algorithmic Trading.V
idyamurthy, Ganapathy (2009). Pairs Trading: Quantitative Methods and Analysis. 1st ed.
Downloads
Published
Issue
Section
How to Cite
