Quantocracy: This is a curated mashup of quantitative trading links.
Reinforcement Learning for Credit Index Option Hedging – Francesco Mandelli, Marco Pinciroli, Michele Trapletti, Edoardo Vittori
In this paper, we focus on finding the optimal hedging strategy of a credit index option using reinforcement learning. We take a practical approach, where the focus is on realism i.e. discrete time, transaction costs; even testing our policy on real market data. We apply a state of the art algorithm, the Trust Region Volatility Optimization (TRVO) algorithm and show that the derived hedging strategy outperforms the practitioner’s Black & Scholes delta hedge.
This study analyzes the transmission of market uncertainty on key European financial markets and the cryptocurrency market over an extended period, encompassing the pre, during, and post-pandemic periods. Daily financial market indices and price observations are used to assess the forecasting models. We compare statistical, machine learning, and deep learning forecasting models to evaluate the financial markets, such as the ARIMA, hybrid ETS-ANN, and kNN predictive models. The study results indicate that predicting financial market fluctuations is challenging, and the accuracy levels are generally low in several instances. ARIMA and hybrid ETS-ANN models perform better over extended periods compared to the kNN model, with ARIMA being the best-performing model in 2018-2021 and the hybrid ETS-ANN model being the best-performing model in most of the other subperiods. Still, the kNN model outperforms the others in several periods, depending on the observed accuracy measure. Researchers have advocated using parametric and non-parametric modeling combinations to generate better results. In this study, the results suggest that the hybrid ETS-ANN model is the best-performing model despite its moderate level of accuracy. Thus, the hybrid ETS-ANN model is a promising financial time series forecasting approach. The findings offer financial analysts an additional source that can provide valuable insights for investment decisions.
Bayesian Forecasting of Stock Returns on the JSE using Simultaneous Graphical Dynamic Linear Models – Nelson Kyakutwika, Bruce Bartlett
Cross-series dependencies are crucial in obtaining accurate forecasts when forecasting a multivariate time series. Simultaneous Graphical Dynamic Linear Models (SGDLMs) are Bayesian models that elegantly capture cross-series dependencies. This study forecasts returns of a 40-dimensional time series of stock data from the Johannesburg Stock Exchange (JSE) using SGDLMs. The SGDLM approach involves constructing a customised dynamic linear model (DLM) for each univariate time series. At each time point, the DLMs are recoupled using importance sampling and decoupled using mean-field variational Bayes. Our results suggest that SGDLMs forecast stock data on the JSE accurately and respond to market gyrations effectively.
Chebyshev polynomials of the first kind have long been used to approximate experimental data in solving various technical problems. Within the framework of this study, the dynamics of shares of eight Czech enterprises was analyzed by the Chebyshev polynomial decomposition: CEZ A.S. (CEZP), Colt CZ Group SE (CZG), Erste Bank (ERST), Komercni Banka (BKOM), Moneta Money Bank A.S. (MONET), Photon (PENP), Vienna insurance group (VIGR) in 2021. An investor, when making a decision to purchase a security , is guided largely by an heuristic approach . And variance and correlation are not observed by human senses. The vectors of decomposition of time series of exchange values of securities allow analyzing the dynamics of exchange values of securities more effectively if their dynamics does not correspond to the normal distribution law. The proposed model allows analyzing the dynamics of the exchange value of a securities portfolio without calculating variance and correlation. This model can be useful if the dynamics of the exchange values of securities does not obey, due to certain circumstances, the normal law of distribution.
Machine learning for option pricing: an empirical investigation of network architectures – Laurens Van Mieghem, Antonis Papapantoleon, Jonas Papazoglou-Hennig
We consider the supervised learning problem of learning the price of an option or the implied volatility given appropriate input data (model parameters) and corresponding output data (option prices or implied volatilities). The majority of articles in this literature considers a (plain) feed forward neural network architecture in order to connect the neurons used for learning the function mapping inputs to outputs. In this article, motivated by methods in image classification and recent advances in machine learning methods for PDEs, we investigate empirically whether and how the choice of network architecture affects the accuracy and training time of a machine learning algorithm. We find that for option pricing problems, where we focus on the Black–Scholes and the Heston model, the generalized highway network architecture outperforms all other variants, when considering the mean squared error and the training time as criteria. Moreover, for the computation of the implied volatility, after a necessary transformation, a variant of the DGM architecture outperforms all other variants, when considering again the mean squared error and the training time as criteria.
Pairs Trading Using Clustering and Deep Reinforcement Learning – Raktim Roychoudhury, Rahul Bhagtani, Aditya Daftari
Conventional pairs trading strategies are based on concepts of mean reversion and stationary stochastic processes, where pairs are assumed to have linear relationships. However, empirical evidence suggest that asset prices in equity markets frequently exhibit non-linear dynamics. We aim to tackle this issue by exploring pairs trading from a deep learning perspective to leverage these non-linear relationships. Our proposed workflow includes two steps, first clustering equity indices and then using a Reinforcement Learning based trading strategy on pairs selected from these clusters. Using a combination of fundamental and technical signals we extract a set of 10 latent risk factors using a convolutional auto-encoder and create 10 clusters of indices using these risk factors. We test our approach on a sample of 13 pairs to train and test our trading strategy and compare our performance against the S&P 500. From the period starting from April 2017 to December 2022, our best performing strategy consisting of the NASDAQ Composite and MSCI World IT Index earns an annualized return of 21.86% with a Sharpe ratio of 1.15, while generating an alpha of 20%.