Forecasting Volatility of Dhaka Stock Exchange: Linear Vs Non-linear models

Masudul Islam, Lasker Ershad Ali, Nahida Afroz


DOI: https://doi.org/10.12777/ijse.3.2.4-8

Abstract


Prior information about a financial market is very essential for investor to invest money on parches share from the stock market which can strengthen the economy. The study examines the relative ability of various models to forecast daily stock indexes future volatility. The forecasting models that employed from simple to relatively complex ARCH-class models. It is found that among linear models of stock indexes volatility, the moving average model ranks first using root mean square error, mean absolute percent error, Theil-U and Linex loss function  criteria. We also examine five nonlinear models. These models are ARCH, GARCH, EGARCH, TGARCH and restricted GARCH models. We find that nonlinear models failed to dominate linear models utilizing different error measurement criteria and moving average model appears to be the best. Then we forecast the next two months future stock index price volatility by the best (moving average) model.


Full Text:

PDF

Refbacks

  • There are currently no refbacks.


Published by Department of Chemical Engineering University of Diponegoro Semarang
Google Scholar

IJSE  by http://ejournal.undip.ac.id/index.php/ijse is licensed under Creative Commons Attribution 3.0 License.