The ETS Challenges: A Machine Learning Approach to the Evaluation of Simulated Financial Time Series for Improving Generation Processes

Abstract

This article presents an evaluation framework that attempts to quantify the degree of realism of simulated financial time series, whatever the simulation method might be, with the aim of discovering and improving unknown characteristics that are not being properly reproduced by such methods. For that purpose, the evaluation framework is posed as a machine learning problem in which the given time series examples must be classified as simulated or real financial time series. The challenge is proposed as an open competition, similar to those published on the Kaggle platform, in which participants must send their classification results and a description of the features and the classifiers used. The results of these challenges have revealed some interesting properties of financial data and have led to substantial improvements in the simulation methods under study, some of which will be described in this work.

TOPICS: Big data/machine learning, performance measurement