Forecasting time-varying daily betas: a new nonlinear approach

Abstract

Purpose -The purpose of this paper is to examine the predictive ability of different well-known models for capturing time variation in betas against a novel approach where the beta coefficient is treated as a function of market return. Design/methodology/approach -Different GARCH models, the Kalman filter algorithm and the Schwert and Seguin model are used against our novel approach. The mean square error, the mean absolute error and the Diebold and Mariano test statistic constitute the measures of forecast accuracy. All models are tested over nine consecutive years and three different samples. Findings -The results show substantial differences in predictive accuracy among the samples. The new approach of modelling the systematic risk overwhelms the rest of the models in longer samples. In the smallest sample, the Kalman filter random walk model prevails. The examination of parameters between two groups of stocks with best and worst accuracy results depicts significant variations. For these stocks, the iid assumption of return is rejected and large differences exist on diagnostic tests. Originality/value -This study contributes to the literature with different ways. First, it examines the predictive accuracy of betas with different well-known models and introduces a novel approach. Second, after constructing betas from the estimated models' parameters, they are used for out-of-sample instead of in-sample forecasts over nine consecutive years and three different samples. Third, a more closely examination of the models' parameters could signal at an early stage the candidate models with the expected lowest forecasting errors. Finally, the study carries out some diagnostic tests for examining whether the existence of iid normal returns is accompanied by better performance.

Testing and comparing conditional risk‐return relationship with a new approach in the cross‐sectional framework

“…The exponential constant term could be thought of being a constant beta coefficient when the remaining terms are zero. Messis and Zapranis (2016) use these specific coefficients for predicting stock betas and compare their accuracy prediction with other well‐known models. The results indicate that the new approach overwhelms the other models in longer samples.…”

Impact of regulatory announcements on systemic risk in the Indian telecom sector

“…This study uses the two step procedure for finding the effect of regulation on systemic risk. The two steps are: The first step was to estimate systemic risk ( β ) using Kalman filter along with CAPM (Barcelos and da Silveira Bueno, 2010; Chalmeau, 2012; Messis and Zapranis, 2016). The β values were not expected to be constant due to the regulatory and policy announcements during the study period, which is confirmed by applying the stationarity tests. The second step was the use of event study methodology for finding the effect of regulatory announcements on daily β (Binder, 1985; Chalmeau, 2012).…”

“…The first step was to estimate systemic risk ( β ) using Kalman filter along with CAPM (Barcelos and da Silveira Bueno, 2010; Chalmeau, 2012; Messis and Zapranis, 2016). The β values were not expected to be constant due to the regulatory and policy announcements during the study period, which is confirmed by applying the stationarity tests.…”