Econometric models for macroeconomic prediction in Argentina

Working papers (original series) | 2001 | N 19

Abstract

The objective of this work is to evaluate the performance of different econometric models—ARIMA and VAR—in generating short and medium-term predictions for some real variables of the economy—GDP, Imports and Investment. In the case of the VAR models, in addition to the variables studied, it was also tried to include MERVAL—as another endogenous variable—in two different ways to try to improve the performance of this type of models. Predictions of one step and two steps forward were then made, which were evaluated using both the absolute prediction error and the root mean square error as a comparison tool.