Nonparametric Test for Financial Time Series Comparisons

The aim of this work concerns the problem of comparing groups of time series, in particular financial time series. Some empirical studies have been published on the topic. However, there is a lack of literature about valid statistical inferential approaches regarding the comparison between groups. In particular, we focus on a two-sample testing problem with the goal of comparing two different groups of financial titles in a given time period. The dataset consists in the time series of the financial returns of the two groups of titles. The problem can be defined as a multivariate test on central tendency and the proposed solution is based on the methodology of combined permutation tests. The application presented in this study concerns the comparative evaluation of the financial performance of ESG titles.