On Pairwise Granger-causality Tests and Econometric Analysis of Selected Economic Indicators: The Case of Nigeria

by Olushina Olawale Awe .

Abstract: The goal of most empirical studies in econometrics and other social sciences is to determine whether a change in one variable causes a change in or helps to predict another variable. Granger causality modeling approach is quite popular in experimental and non-experimental fields which involve some dynamic econometric time series methodologies. In this paper, Granger causality and co-integration tests were employed in the empirical modelling of seven economic indicators in Nigeria. The results alternated between bi-directional, uni-directional and no causality among the economic indicators considered. Prior to the Granger causality tests, we tested for stationarity in the variables using the Augmented Dickey-Fuller (ADF) procedure. The variables proved to be integrated of either I(1) or I(2). Johansen co-integration test reveals that at 5% level of significance, we have at least four co-integrating pairs among the variables. This verifies the fact that when two or more time series are co-integrated, there must be either bi-directional or uni-directional Granger causality between them. Our findings reveal that Government Investment, Real Money Supply and Government Expenditure Granger causes output growth in Nigeria. Finally, we relate these results with popular postulations in economic theory.

Key Words: Causality, GDP, Co-integration, Prediction, Economic theory

Author:
Olushina Olawale Awe, olawaleawe@yahoo.co.uk

Editor: Mark Greenwood, greenwood.stat@gmail.com

READING THE ARTICLE: You can read the article in portable document (.pdf) format (673180 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 1160 times since August 17, 2012.


Return to the InterStat Home Page.