[R-sig-ME] mixed probit and rare events (MCMCglmm)

[Linus Holtermann]

Hello,



I like to estimate a mixed model via MCMCglmm. My model is a binomial model (probit). There are more than 3000 observations in my data, but for some investigations the dependent variable is one for only 320 observations (zero otherwise). Is this a problem in a MCMC setup? How can I solve this problem of rare events in a probit model? Note, I want to stick to MCMC. Is a Gelman-Prior a proper choice? Are there other solutions that are implemented in MCMCglmm (or in other packages)?

Best regards,

Linus Holtermann
Hamburgisches WeltWirtschaftsInstitut gemeinnützige GmbH (HWWI)
Heimhuder Straße 71
20148 Hamburg
Tel +49-(0)40-340576-336
Fax+49-(0)40-340576-776
Internet: www.hwwi.org<http://www.hwwi.org/>
Email: holtermann at hwwi.org<mailto:holtermann at hwwi.org>

Amtsgericht Hamburg HRB 94303
Geschäftsführer: Prof. Dr. Henning Vöpel
Prokura: Dipl. Kauffrau Alexis Malchin
Umsatzsteuer-ID: DE 241849425


	[[alternative HTML version deleted]]