Step 1 of 3. This form allows you to add a new citation to our database.
Our data says that the document:
Majda, Benzidia ; Michel, Lubrano ; Michel, Lubrano ; Majda, Benzidia. (2016) A Bayesian Look at American Academic Wages: The Case of Michigan State University.
In: AMSE Working Papers. RePEc:aim:wpaimx:1628.
Full description at Econpapers
cites:
log(xi/h). Bayesian inference, as provided by Arnold (2008) requires a Gibbs sampler. As a matter of fact, the Pareto process does not belong to the exponential family, but conditionally on h or conditionally on α, it does. So it is possible to nd natural conjugate priors for α and h, provided we write the likelihood function in the following form: L(x; α, h) = αn exp n −(α + 1) X log(xi) + α n log(h) o 1(x(1) > h).
but, as far as we know, such cited document is not available on RePEc.
If you know this information is not correct, you may help us to improve our service giving us the handle of the cited document in the box below. In this way, we could add the citation to CitEc.
Note that if you are the author of a cited document which it is not available on RePEc, you can submit a copy to the Munich Personal RePEc Archive (MPRA). Upload a paper now.