MODELLING ABRUPT SHIFT IN TIME SERIES USING INDICATOR VARIABLE: EVIDENCE FROM NIGERIAN INSURANCE STOCK

No Thumbnail Available
Date
2015-07
Authors
AGBOOLA, Samson
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Abstract This study models abrupt shift in time series using indicator variable: evidence from Nigeria Insurance stocks. Data on daily closing prices for some selected Nigerian Insurance stocks were collected between 2nd January, 2000 and 26th May, 2014. Daily returns were then computed from thedaily prices. To study the volatility pattern of Insurance stock, seven symmetric models and five asymmetric models with dummy variables incorporated to their variance equation were considered and these are ARCH (1), ARCH (2), ARCH (3), GARCH (1, 1), GARCH (2, 1), GARCH (1, 2), GARCH (2, 2), EGARCH (1, 1), EGARCH (1, 2), EGARCH (2, 1), EGARCH (2, 2) and TARCH (1, 1). Post estimation and performance evolution metric was evaluated using the RMSE, MAE and MAPE. The results showed that, the daily returns were stationary but not normally distributed and eight out of ten stocks considered for the study showed evidence of ARCH effect. Furthermore, the results of the post estimation revealed that most of the models were competitive and model ARCH (1) and EGARCH (1, 1) proved to be the most suitable among the twelve competing volatility models considered in some of the Insurance company. This present study findings are very crucial and informative to investors and intending investors as it will help in stock pricing strategy as volatility is the major index used to evaluate asset performance and stock pricing strategy.
Description
A THESIS SUBMITTED TO THE POSTGRADUATE SCHOOL, AHMADU BELLO UNIVERSITY, ZARIA NIGERIA IN PARTIAL FULFILLMENT FOR THE AWARD OF MASTER OF SCIENCE (M.Sc) DEGREE IN STATISTICS DEPARTMENT OF MATHEMATICS AHMADU BELLO UNIVERSITY, ZARIA NIGERIA
Keywords
MODELLING, ABRUPT, SHIFT, INDICATOR, VARIABLE, EVIDENCE, INSURANCE, STOCK
Citation
Collections