Open Science Research Excellence
%0 Journal Article
%A Katleho D. Makatjane and  Kalebe M. Kalebe
%D 2018 
%J  International Journal of Social, Behavioral, Educational, Economic, Business and Industrial Engineering
%B World Academy of Science, Engineering and Technology
%I International Science Index 141, 2018
%T Modelling Conditional Volatility of Saving Rate by a Time-Varying Parameter Model
%U http://waset.org/publications/10009468
%V 141
%X The present paper used time-varying parameters which are based on the score function of a probability density at time t to model volatility of saving rate. We used a scaled likelihood function to update the parameters of the model overtime. Our results revealed high diligence of time-varying since the location parameter is greater than zero. Furthermore, we discovered a leptokurtic condition on saving rate’s distribution. Kapetanios, Shin-Shell Nonlinear Augmented Dickey-Fuller (KSS-NADF) test showed that the saving rate has a nonlinear unit root; therefore, it can be modeled by a generalised autoregressive score (GAS) model. Additionally, value at risk (VaR) and conditional tail expectation (CTE) indicate that 99% of the time people in Lesotho are saving more than spending. This puts the economy in high risk of not expanding. Therefore, the monetary policy committee (MPC) of Lesotho should revise their monetary policies towards this high saving rates risk.

%P 1170 - 1173