Una revisión crítica de los métodos de series de tiempo y de datos de panel aplicados al caso de la hipótesis de export-led growth

ABSTRACTThe main objective of this document is to show the application of modern econometric techniques, and compare the results obtained from non-stationary time series technique against those from the non-stationary panel date technique applied to a specific macroeconomic problem such as the export-led growth hypothesis.From the results of the country-by-country analysis we found evidence in favor of the export-led growth hypothesis only for 5 countries out of 40. On the other hand, using panel data techniques we found that the hypothesis holds for our sample. However, this result is valid once we impose a homogeneity assumption on the cointegration coefficients. These findings suggest caution, at least in the derivation of political economy recommendations respect to the export-led growth hypothesis.