MODELING NIGERIAN STOCK PRICE VOLATILITY USING EGARCH-X MODEL WITH DIFFERENT INNOVATIONS

Abstract

This study explores the modelling performance of EGARCH-X using the skewed student’s t,
normal, and student’s t innovations. The aim of the study was to determine the innovation that
best captures the asymmetry and kurtosis exhibited by the returns on financial data. The
descriptive statistics revealed that the distributions of returns on the stock prices were skewed
and leptokurtic. The unit root test was carried out using the Augmented Dickey-Fuller (ADF)
test. The result of the unit root test reveals that the returns on the series were stationary. The
ARCH LM-test detected the presence of ARCH effects. The mean equation was estimated, and
the EGARCH-X (1,1) model was fitted to the data, incorporating three exogenous variables
(daily opening price, daily low price, and daily high price). The goodness of fit of the models
was tested using Akaike Information Criterion, Bayesian Information Criterion, and Log-
Likelihood. The models' performance, based on Akaike Information Criterion (AIC), Bayesian
Information Criterion (BIC), and Log-Likelihood, revealed that EGARCH-X (1,1) with skewed
student's t innovation performs better than EGARCH-X (1,1) with normal and student’s t
innovations. The findings of the study further revealed that volatility persists longer in the
models with Student’s t innovations, suggesting a slower mean-reverting process pertinent for an
appropriate forecast of the financial market.