Inference and Thick Tails: Some Surprising Results
This paper analyses the distribution of the classical t-ratio statistic from distributions with no finite moments and shows how classical testing is affected. Some surprising results are obtained in terms of bimodality vs. the usual unimodality of the standard studentized t-distribution prevailing in classical conditions. The paper develops a new distribution termed the “double Pareto,” which allows the thickness of the tails and the existence of moments to be determined parametrically. We also consider infinite-moments distributions truncated on a compact support to investigate the relative importance of tail thickness in case of finite moments. We find that the bimodality persists even in such cases. Simulation results are used to highlight the dangers of relying on naive testing in the face of thick-tailed distributions. Special cases analyzed include one- and two-sample statistical inference problems, as well as linear regression econometric problems.
|Date of creation:||03 May 2007|
|Date of revision:|
|Contact details of provider:|| Postal: Via Conservatorio 7 - 20122 Milano|
Phone: +39 02 503 16486
Fax: +39 02 503 16475
Web page: http://services.bepress.com/unimi
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:bep:unimip:unimi-1054. 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: (Christopher F. Baum)
If references are entirely missing, you can add them using this form.