Volatility assessment and Var (P) Model for listed insurance companies stock returns, an overview of Ghana stock exchange market
No Thumbnail Available
Date
April, 2016
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The study investigated the presence of volatility in two insurance companies listed
on the Ghana Stock Exchange as well as model the interdependencies between
these two insurance companies. In checking the presence of volatility, several
ARIMA (p,d,q) models were fitted separately to the log return series of the
two companies and the best model selected using the AIC and BIC selection
criterion. Using the selected ARIMA (p,d,q) models, the residuals of the model
were obtained and presence of ARCH effects evaluated using the ARCH-LM test
for each company. The ARCH-LM test revealed that there were no ARCH effects.
The ARIMA (1,0,2) model was thus used in forecasting the returns of EGL
whereas the ARIMA (1,0,1) model was used in forecasting returns of SIC. Finally,
VAR (p) models were fitted to the combined series to check for interdependencies.
Using the AIC and BIC, the VAR (7) model residuals were found to satisfy the
null hypothesis of no serial correlation between the stocks, thus was selected
as the best model. Investment with SIC and or EGL is strongly recommended
since volatility is absent in their stock prices. The Ghana Stock Market is still
developing, as a consequence management of these two listed companies is advised
to put in place measures that will increase the frequency of trading these stocks.
Description
A thesis submitted to The Department of Mathematics, Kwame Nkrumah University of Science and Technology in partial fulfillment of the requirements for the degree of MSc. Actuarial Science,