Vol. 11 No. 1-2 (2017)
Articles

Comparing parametric and semiparametric error correction models for estimation of long run equilibrium between exports and imports

Published June 30, 2017
Henry De-Graft Acquah
University of Cape Coast, Cape Coast, Ghana, Department of Agricultural Economics and Extension
Lawrence Acheampong
Department of Agricultural Economics and Extension, University of Cape Coast, Cape Coast, Ghana
pdf

APA

De-Graft Acquah, . H. ., & Acheampong, L. . (2017). Comparing parametric and semiparametric error correction models for estimation of long run equilibrium between exports and imports. Applied Studies in Agribusiness and Commerce, 11(1-2), 19–23. https://doi.org/10.19041/APSTRACT/2017/1-2/3

This paper introduces the semiparametric error correction model for estimation of export-import relationship as an alternative to the least squares approach. The intent is to demonstrate how semiparametric error correction model can be used to estimate the relationship between Ghana’s export and import within the context of a generalized additive modelling (GAM) framework. The semiparametric results are compared to common parametric specification using the ordinary least squares regression. The results from the semiparametric and parametric error correction models (ECM) indicate that the error correction term and import variable are significant determinants of Ghana’s exports. On the basis of Akaike Information Criteria and Generalized Cross-Validation (GCV) scores, it is found that the semiparametric error correction model provides a better fit than the widely used parametric error correction model for modeling Ghana’s export-import relationship. The results of the analysis of variance provide further evidence of nonlinearity in Ghana’s export and import relationship. In effect, this paper demonstrates the usefulness of semiparametric error correction model in the estimation of export – import relationship.

JEL code: C14, C18, C22, F10, F14