Functional Analysis of Variance: An Application to Stock Exchange

Abstract

The concept of "functional data" allows for the representation of data collected repeatedly over a period of time as a continuous function within a specific range on the time axis, rather than as discrete measurement points. Traditional statistical analysis has been adapted to accommodate functional data. This paper discusses the adaptation of one-way analysis of variance for functional data, covering parameter estimations and obtaining test statistics. As a numerical example, stock exchange values from various countries across different continents are used. The aim is to discern potential differences among countries based on these stock exchange values during the Covid-19 pandemic, utilizing one-way analysis of variance for functional data.