Title:Multivariate Gaussian and Student-t Process Regression for Multi-output Prediction and Stock Market Modelling

Abstract:Gaussian process for vector-valued function model has been shown to be a useful method for multi-output prediction. The existing method for this model is to re-formulate the matrix-variate Gaussian distribution as a multivariate normal distribution. Although it is effective and convenient in many cases, re-formulation is not always workable and difficult to extend because not all matrix-variate distributions can be transformed to related multivariate distributions, such as the case for matrix-variate Student-t distribution. In this paper, we propose the multivariate Gaussian process regression (MV-GPR) and Student-t process regression (MV-TPR) for multi-output prediction, where the model settings, derivations and computations are all directly performed in matrix form, rather than vectorizing the matrices involved in the existing method of Gaussian process for vector-valued function model. The effectiveness of the proposed methods are illustrated through several simulated examples. The proposed methods are then applied to stock market modelling. The Buy&Sell strategies, which are generated by our proposed methods, are shown to be profitable in the equity investment.