Title: Improvements in forecasting of bank stock excess returns using the investor sentiment endurance index: a comparison with CAPM and Fama-French models

Authors: Ling T. He; K. Michael Casey

Addresses: Department of Economics and Finance, University of Central Arkansas, Conway, AR 72035, USA ' Department of Economics and Finance, University of Central Arkansas, Conway, AR 72035, USA

Abstract: In order to raise the forecasting quality for banking equity costs, this study uses the sentiment endurance (SE) index developed by He (2012) and applies this model to the banking industry. The SE index is used as the risk factor to replace commonly used risk factors, the overall market risk premium, and the Fama-French factors 'small minus big' (SMB), and 'high minus low' (HML). The sentiment endurance index in this study measures changes in the lasting momentum of bank stock prices and can therefore be used to predict future changes in bank stock prices. The results of this study indicate that the monthly rolling out-of-sample forecasts generated by the sentiment endurance model, in general, are significantly more accurate than the CAPM and Fama-French models. When the overall market risk premium or SMB and HML are added into the sentiment endurance index model, respectively, the quality of forecasting based on short rolling periods actually deteriorates and improvements in forecasting quality based on longer rolling periods are trivial. The empirical results indicate that information contained in the three risk factors is already reflected in the sentiment endurance index.

Keywords: investor sentiment endurance index; rolling forecasting.

DOI: 10.1504/IJFMD.2018.091679

International Journal of Financial Markets and Derivatives, 2018 Vol.6 No.3, pp.210 - 224

Received: 17 Jun 2017
Accepted: 22 Aug 2017
Published online: 11 May 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article