In bayesian analysis problem, The priors of the truncated normal distribution at zero distribution's parameters (location and scale parameters), I need to find (if exist) the informative priors. I read some papers about this topic, but they could not solve it (in MCMC algorithm, it was difficult).
Can you clear up the question ? Prior distributions are derived from previous knowledge or are just chosen, according to reasonable principles, such as not excluding any value, or Cauchy distributions (without finite expectation, but with finite 50% value), or a Uniformly Most Powerful Bayesian test (see Johnson pnas, 2013, 110, 48, 19313-19317)
.
i suppose you're rather looking for "non-informative" priors ... if interested in "informative" priors, you should first consider the kind of prior information you want to embed in your prior !
.
hunting for a (somewhat) conjugate prior ? the paper attached treats the half normal ; quite ugly ...
.
the link below gives the form of the Jeffrey's prior associated to the truncated normal
http://irnd.org/papers/TruncatedData_R2.pdf
.
Who has the full-text article with title of "Approximate formulas for the confidence intervals of process capability indices" writen by Nagata, Y. and Nagahata, H. (1992)?
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics