GRANGER CAUSALITY FROM THE EXCHANGE RATE TO AGRICULTURAL PRICES AND EXPORT SALES

In-sample and out-of-sample Granger causality tests are applied to determine whether the real trade-weighted agricultural exchange rate helps predict monthly real prices and export sales of wheat, corn, and soybeans. An ARIMA model, alternative univariate and bivariate autoregressive models, and a restricted bivariate autoregressive model based on Hsiao's procedure are specified for each variable. Results of the causality are shown to be sensitive to specification choice. Forecasting performance of the models is compared and out-of-sample Granger causality is determined using univariate and bivariate models with the best (lowest mean-square forecast error) forecasting accuracy. These tests provide evidence supportive of Granger causality from the exchange rate to export sales, while the evidence on causality from the exchange rate to prices is mixed.


Issue Date:
1990-07
Publication Type:
Journal Article
DOI and Other Identifiers:
Record Identifier:
https://ageconsearch.umn.edu/record/32503
PURL Identifier:
http://purl.umn.edu/32503
Published in:
Western Journal of Agricultural Economics, 15, 1
Page range:
100-110
Total Pages:
11




 Record created 2017-04-01, last modified 2019-08-26

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)