How to determine priors?

[Saurabh]

Hello People, I have developed a model with many new kind of frictions. Due to lack of literature available currently, i just want a general guidance on how to determine the priors of persistence and standard devaition of these new shocks?

Thank you

[jpfeifer]

Please see the discussion at

DSGE Estimation - Imposing prior mean and std

Please see [Choosing an appropriate standard deviation for prior) and if this does not answer your question, report back. Formal prior elicitation is done in Del Negro Schorfheide (2008) - Forming priors for DGSE models. Regarding the error message: you are using a beta distribution that implies the parameter is between 0 and 1. Given that no realizations can be bigger than 1 or smaller than 0, this limits the standard deviation that is possible. If you impose a mean of 0.86, the parameter can …

[Saurabh]

Thank you Professor!!!

[Saurabh]

Hey, I am using 13 observed variables. For each observed variable, I have 24 observations at annual frequency. I have used one-sided HP filter on my data.

Total number of structural shocks are 10. In addition to it, I have included measurement errors as well.

Just want to know, Is it OK, what I am doing?

[jpfeifer]

Just based on your description, there is nothing here that is immediately wrong.

[Saurabh]

I wonder if the length of time series is sufficient to produce credible results?

[jpfeifer]

This is not a concept applicable for a Bayesian. If the sample is short or the data informative, your prior will not be updated by much. But that is perfectly fine. You just need to openly communicate this.