Scale-dependence of the Negative Binomial Pseudo-Maximum Likelihood Estimator
Following Santos Silva and Tenreyro (2006), various studies have used the Poisson Pseudo-Maximum Likelihood to estimate gravity specifications of trade flows and non-count data models more generally. Some papers also report results based on the Negative Binomial estimator, which is more general and encompasses the Poisson assumption as a special case. This note shows that the Negative Binomial estimator is inappropriate when applied to a continuous dependent variable which unit choice is arbitrary, because estimates artificially depend on that choice.
|Date of creation:||2010|
|Date of revision:|
|Note:||View the original document on HAL open archive server: https://halshs.archives-ouvertes.fr/halshs-00535594|
|Contact details of provider:|| Web page: https://hal.archives-ouvertes.fr/|
Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.:
- Joao Santos Silva & Silvana Tenreyro, 2005.
"The log of gravity,"
LSE Research Online Documents on Economics
3744, London School of Economics and Political Science, LSE Library.
- Lionel Fontagne & Soledad Zignago, 2007.
"A Re-evaluation of the Impact of Regional Agreements on Trade Patterns,"
CEPII research center, issue 109, pages 31-51.
- Lionel Fontagné & Soledad Zignago, 2007. "A Re-evaluation of the impact of regional agreements on trade patterns," Université Paris1 Panthéon-Sorbonne (Post-Print and Working Papers) hal-00270499, HAL.
- Joakim Westerlund & Fredrik Wilhelmsson, 2009. "Estimating the gravity model without gravity using panel data," Applied Economics, Taylor & Francis Journals, vol. 43(6), pages 641-649.
When requesting a correction, please mention this item's handle: RePEc:hal:cesptp:halshs-00535594. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (CCSD)
If references are entirely missing, you can add them using this form.