FURTHER DISCUSSION PURSUANT TO MY DEBATE WITH JOHN K. KRUSCHKE

                                                                                             

                                                                                PREVIOUS DEBATE

Thomas Leonard From my blog:THIS ITEM WAS RECENTLY PUBLISHED BY JOHN KRUSCHKE ON THE JUNIOR ISBA FACEBOOK PAGE, with a readership of over 1000 Bayesian Statisticians. Fixing seems to be an appropriate word. The 95% HDI later referred to by John may, or may not, have been based upon non-standard assumptions, My understanding, from John's later discussion of 'heavy tails', is that it is much too broad when compared with the standard procedures. The state of the art concerning the Bayesian investigation of hypotheses is described by BASKURT AND EVANS in their recent paper in Bayesian Analysis. In regression situations where there is vague prior knowledge and under appropriate normality assumptions, their procedures are consistent with the usual F, chi-squared, and t-tests, but with weights of evidence which supplement the usual p-values. John seems to be unaware of this.

Like · Reply · 20 hrs

Vini Fonz As stated, he was just showing, with a simple example, how to fix the intercept at zero in jags. 
If I'm not wrong in he's books he works with those concepts you're saying he seems to be unaware of. 
Nonetheless I got some very good information about real life procedures from your coments Leonard. Thank you guys for sharing all this knowledge with us!

Unlike · Reply · 1 · 19 hrs · Edited

Thomas Leonard Thank you, I believe however that regression by MCMC should only be employed after the data have been analysed using a standard regression analysis, and then only if there are good practical reasons e.g, the availability of prior information, for using...See More

Like · Reply · 1 · 7 hrs

Vini Fonz Yes I understand that. That example works more like the other examples on Kruschke's book, they are simple and not very intensive coz they are educational. The focus is on how to use and tune the tools.

Like · Reply · 7 hrs

Vini Fonz About the heavy tailed distributions. In my area, Ecology, more specifically, marine ecology, the data is more often than not, heavy tailed. It's the expected for things like species richness, individuals counts, biomass per unit of effort, etc, etc. A...See More

Like · Reply · 7 hrs

Thomas Leonard in this case generalised linear models are extremely valuable, and they can be analysed using the Laplacian approximations employed in the INLA package,

Like · Reply · 1 · 7 hrs

Vini Fonz Exactly, I just finished my master's degree last week and it was all spatial models with INLA. 
Do you think it will be possible in the future to estimate joint distributions using Laplacian approximations either in INLA or other methods?

Like · Reply · 6 hrs · Edited

Thomas Leonard These applications of Laplacian methods are discussed more fully in my book Bayesian Methods (Cambridge University Press, with John Hsu). Your Masters degree sounds very interesting, Who were your mentors?

Like · Reply · 12 mins

                                                               Vini Fonz Well, I'm looking into buying your book now, looks like a 

great reading. I'm quite a beginner in Bayesian statistics, so I'm trying to get as much knowledge as I can. 
My mentor, who introduced me to Bayesian statistics and INLA was Maria Grazia Pennino, she has several nice publications related to fisheries and Bayesian methods:

https://scholar.google.com/citations?user=1T-UWYsAAAAJ...

Maria Grazia Pennino - Google Scholar Citations

This "Cited by" count includes citations to the following articles in Scholar. The ones marked * may be different from the article in the profile.

SCHOLAR.GOOGLE.CO.UK

Unlike · Reply · 1 · 1 hr

Write a reply...

Write a comment...

BAYESIAN METHODS: AN ANALYSIS FOR STATISTICIANS AND INTERDISCIPLINARY RESEARCHERS