« Chapter 6 ANOVA models Ntzoufras | Main | Chill your wet ipad in the fridge to dry it »

July 15, 2014

How to get shape and rate constants for gamma pseudopriors

Kruschke, J. K. (2011). Doing Bayesian Data Analysis explains how to get constants for gamma pseudo priors. I got the idea. I can also use the mean and standard deviation of a sample of  kappa values to get the shape and rate constants. (p. 251)

The problem is,  I am not a stat genius. I am just a sociologist! I do not know R very well, either.  I had to think how I get the mean and s.d. from the first run that uses the real priors. I write it here so that I can go back if I forget. And I am pretty sure I will forget.

The sample program stores kappa values in kappa1sampleM2, kappa3sampleM2, (i.e. kappa[1]model2. the number after kappa refers to the condition number)....etc.
So it seems I have to get a mean and s.d, like this:

kappamean <- mean(kappa1sampleM2)
kappasd     <- sd(kappa1sampleM2)


« Chapter 6 ANOVA models Ntzoufras | Main | Chill your wet ipad in the fridge to dry it »

Comments

Post a comment



(Not displayed with comment.)


Comments are moderated, and will not appear on this weblog until the author has approved them.



« Chapter 6 ANOVA models Ntzoufras | Main | Chill your wet ipad in the fridge to dry it »