Markov switching regression with interval data: Application to financial risk via CAPM

Abstract

© 2017 American Scientific Publishers. All rights reserved. In the past, the study of finance have basically focused on single-valued data which may not well represent the stock price behaviour. Therefore, this paper suggests the approach that gains more efficiency from using the interval-valued data. Moreover, due to the nonlinear behavior of the financial data, we apply a Markov switching approach to interval-valued data and propose the Markov switching interval regression. We apply this approach to the capital asset pricing model or CAPM and introduce the Markov switching CAPM with interval-valued data as the originality of this paper. We, then, apply our model to the real stock price intervals, namely highest and lowest prices. The overall results suggest that our proposed model can perform very well and it is also able to capture a nonlinear behavior of the stock at the same time.