FORECASTING THE TIME VARYING-BETA OF NSE-20 SHARE COMPANIES: BI-VARIATE GARCH (1, 1) MODEL VS KALMAN FILTER METHOD.

This research paper forecasts the time-varying weekly beta of ten stocks listed in the Nairobi Securities Exchange 20-Share Index by Bivariate GARCH (1, 1) and Kalman filter methods. A comparison of the forecasting ability of the GARCH model and the Kalman filter method is made. Forecast errors based on the returns on assets are used to evaluate the out-of-sample forecast ability of both methods. Two measures, Mean Absolute Error (MAE) and Mean Squared Error (MSE). The results about which method is superior over the other is inconclusive .Comparing the two, the Kalman method is superior over the Bivariate GARCH (1 , 1) based on MSE while the Bivariate GARCH (1 ,1) provide more accurate forecast of the time varying beta.

 

Key words: Bivariate GARCH (1, 1), Forecasting, Kalman Filter.

 

Documentation : 

Message From the Director

Workshops and Seminars

Community Outreach

Contact Us

School of Mathematics, CBPS College.

P. O. Box 30197 - 00100
Tel: 254-02-4445751.

Mobile no :0780-834766.
Email: maths@uonbi.ac.ke.

Facebook

twitter:@uon_maths

 

UoN Website | UoN Repository | ICTC Website


Copyright © 2019. ICT WebTeam, University of Nairobi