A0471
Title: GMANOVA modelling for volatile data
Authors: Sayantee Jana - Indian Institute of Technology Hyderabad (India) [presenting]
Abstract: Generalized multivariate analysis of variance (GMANOVA) models are linear models useful for the analysis of longitudinal data, which are repeated measurements of a continuous variable from several individuals across any ordered variable such as time, temperature, pressure, etc. GMANOVA models are widely used in economics, social sciences and medical research. However, despite financial data being time-varying, the traditional GMANOVA model has limited to no applications in finance due to the volatile nature of such data. This, in turn, makes financial data the right candidate for Multivariate t (MT) distribution, as it allows for outliers in the data to be modelled due to its heavy tails. In fact, portfolio analysis, including mutual funds and capital asset pricing, are all modelled using elliptical distributions, especially MT distribution. The classical GMANOVA model assumes multivariate normality, and hence, the inferential tools developed for the classical GMANOVA model may not be appropriate for heavy-tailed data. The sensitivity of inferential tools developed under multivariate normality under volatile data is first explored, and then inferential tools are developed for the GMANOVA model under the MT distribution. The practical implementability of the proposed method is demonstrated on a financial dataset.
