Model choice and size distribution: a Bayequentist approach
Publikation: Arbeits- oder Diskussionspapiere und Berichte › Arbeits- oder Diskussionspapiere
Standard
Lüneburg: Institut für Volkswirtschaftslehre der Universität Lüneburg, 2013. (Working Paper Series in Economics; Nr. 265).
Publikation: Arbeits- oder Diskussionspapiere und Berichte › Arbeits- oder Diskussionspapiere
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - UNPB
T1 - Model choice and size distribution: a Bayequentist approach
AU - Engler, John-Oliver
AU - Baumgärtner, Stefan
PY - 2013/2/28
Y1 - 2013/2/28
N2 - We propose a new three-step model-selection framework for size distributionsin empirical data. It generalizes a recent frequentist plausibility-of-t analysis(Step 1) and combines it with a relative ranking based on the Bayesian AkaikeInformation Criterion (Step 2). We enhance these statistical criteria with theadditional criterion of microfoundation (Step 3) which is to select the sizedistribution that comes with a dynamic micro model of size dynamics. Anumerical performance test of Step 1 shows that our generalization is ableto correctly rule out the distribution hypotheses unjustied by the data athand. We then illustrate our approach, and demonstrate its usefulness, with asample of commercial cattle farms in Namibia. In conclusion, the frameworkproposed here has the potential to reconcile the ongoing debate about sizedistribution models in empirical data, the two most prominent of which arethe Pareto and the lognormal distribution.
AB - We propose a new three-step model-selection framework for size distributionsin empirical data. It generalizes a recent frequentist plausibility-of-t analysis(Step 1) and combines it with a relative ranking based on the Bayesian AkaikeInformation Criterion (Step 2). We enhance these statistical criteria with theadditional criterion of microfoundation (Step 3) which is to select the sizedistribution that comes with a dynamic micro model of size dynamics. Anumerical performance test of Step 1 shows that our generalization is ableto correctly rule out the distribution hypotheses unjustied by the data athand. We then illustrate our approach, and demonstrate its usefulness, with asample of commercial cattle farms in Namibia. In conclusion, the frameworkproposed here has the potential to reconcile the ongoing debate about sizedistribution models in empirical data, the two most prominent of which arethe Pareto and the lognormal distribution.
KW - Economics
KW - Nachhaltigkeitsökonomie
KW - Sustainability Economics
M3 - Working papers
T3 - Working Paper Series in Economics
BT - Model choice and size distribution: a Bayequentist approach
PB - Institut für Volkswirtschaftslehre der Universität Lüneburg
CY - Lüneburg
ER -