[WWW19] Linear mixed-effects model for longitudinal complex data with diversified characteristics

Revue Internationale avec comité de lecture : Journal Journal of Management Science and Engineering, pp. 1-28, 2019, (doi:

Mots clés: Longitudinal complex data,Linear mixed-effects model,Compositional data analysis,Functional data analysis, Stock market, Online investors’ emotions

Résumé: The increasing richness of data encourages a comprehensive understanding of economic and financial activities, where variables of interest may include not only scalar (point-like) indicators, but also functional (curve-like) and compositional (pie-like) ones. In many research topics, variables are also chronologically collected across individuals, which falls into the paradigm of longitudinal analysis. The complicated nature of data, however, increases the difficulty of modeling these variables under a traditional longitudinal framework. In this study, we investigate a linear mixed-effects model (LMM) for such complex data. Different types of variables are first consistently represented using the corresponding basis expansions so that the LMM can then be conducted on them, which generalizes the theoretical framework of the LMM to complex data analysis. A number of numerical experiments indicate the feasibility and effectiveness of the proposed model. We further illustrate its practical utility in a real data study of China’s stock market and show that the proposed method can enhance the performance and interpretability of the regression for complex data with diversified characteristics.


@article {
title="{Linear mixed-effects model for longitudinal complex data with diversified characteristics}",
author="z. wang and H. Wang and S. Wang and S. Lu and G. Saporta",
journal="Journal of Management Science and Engineering",