Modelling international trade data with the Tweedie distribution
Robust methods are indispensable in the analysis of international trade data, as outliers and deviations from the classical model assumptions are recurring issues. The objective evaluation of such methods requires a large number of data sets articially generated with known statistical properties, which must comply with real world scenarios. This contribution shows the potential of the Tweedie distribution (Tweedie , 1984) in mimicking the distribution of the quantities of given products imported into the European Union.
BARABESI Lucio;
CERASA Andrea;
PERROTTA Domenico;
CERIOLI Andrea;
2014-10-20
Martin-Luther-University Halle-Wittenberg
JRC90373
Additional supporting files
| File name | Description | File type | |
|---|
