Document Type

Article

Journal/Book Title/Conference

Economics Research Institute Study Paper

Volume

95

Issue

6

Publisher

Utah State University Department of Economics

Publication Date

1995

First Page

1

Last Page

28

Abstract

The introduction of generalized autoregressive conditional heteroscedastic (GARCH) modelling techniques to agricultural price analysis represents an important advancement, both because commodity prices often exhibit volatility clustering and because explicit estimation of conditional second moments is desirable where price risk influences production and marketing behavior. However, agricultural economists may be borrowing too freely from the macro econometrics literature in applying parsimonious GARCH models to commodity price series. Using monthly data from Madagascar, this paper demonstrates the value of introducing structural regressors to the conditional variance equation of GARCH models of commodity prices. The extended GARCH model is shown to exhibit superior out-of-sample forecasting accuracy.


Download

Share

COinS