search for


Comparative study between functional data analysis and multivariate data analysis for functional data
Journal of the Korean Data & Information Science Society 2022;33:817-27
Published online September 30, 2022;
© 2022 Korean Data and Information Science Society.

Kyungmin Ahn1

1Department of Statistics, Keimyung University
Correspondence to: This research was supported by the Bisa Research Grant of Keimyung University in 2022.
1 Assistant professor, Department of Statistics, Keimyung University, Daegu 42601, Korea. E-mail:
Received July 15, 2022; Revised August 4, 2022; Accepted August 13, 2022.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Functional data analysis (FDA) is a branch of statistics that deals with functional variables. In fact, the most general space for functional data is the Hilbert space that it is general to define functional statistical methods in Hilbert structure. Therefore, to analyze the functional data, it is common to apply functional statistical models rather than the models from multivariate data analysis (MDA). As the results, applications of statistical methods from MDA and FDA give very different results in prediction as well as performance. Especially for functional data, FDA always leads to a superior performance. However, there is no paper which compares the difference in performance between MDA and FDA using functional data. Hence, in this paper, we compare the prediction results of MDA and FDA by applying each regression model for functional data.
Keywords : Functional data analysis, functional linear regression model, multiple linear regression model, multivariate data analysis.