Bayesian analysis entails the formulation of a prior distribution for the parameter concerned. On the basis of available data the posterior distribution of the parameter is obtained. If the loss function adopted is quadratic the Bayesian solution is the posterior mean. If the loss function is the absolute function the Bayesian solution is the posterior median.

Therefore using the method of moments with the quadratic loss function the sample mean is an estimate of the parameter. With the absolute loss function the sample median estimates the parameter.

I hope this contribution will be of assistance in the coding.

Thanks for your response

Actually my model is following

GDP = response variable and (corruption, capital development, financial development,trade openness ) are regressors.

I have estimate the model through classical Generalized method of moment in Rstudio software. Now I want to estimate the model through Bayesian Generalized Method of Moment.

I think there is no one who work on Bayesian generalized method of moment, if any one can understand the this endogeneitic problem of the model please guide me.

Dear Muhammad,

regarding your initial question, I recommend that you take a look at the following threat, where implementation is discussed based on the Yin Paper from 2009.

https://groups.google.com/forum/#!topic/stan-users/TWsGQytpYLQ

I also recommend to read the Paper from Yin, since the actual method is explained in the framework of generalized linear (mixed) models.

Complex Bayesian modelling is not part of my expertise, therefore I recommend you to get in contact with Mr Yin, who is now Professor at the University of Texas and provides software to download. http://web.hku.hk/~gyin/ ; software: https://biostatistics.mdanderson.org/SoftwareDownload/