Calcite | Level 5

## Priori hypothesis

How to do priori hypothesis for anova program
4 REPLIES 4
SAS Super FREQ

## Re: Priori hypothesis

Hello,

Are you talking about Bayesian methods?

SAS provides two avenues for Bayesian analysis: built-in Bayesian analysis in certain modeling procedures and the MCMC procedure for general-purpose modeling. The built-in Bayesian procedures are ideal for data analysts beginning to use Bayesian methods, and they suffice for many analysis objectives.

See :
Paper SAS400-2014
An Introduction to Bayesian Analysis with SAS/STAT® Software
Maura Stokes, Fang Chen, and Funda Gunes
SAS Institute Inc.
https://support.sas.com/resources/papers/proceedings14/SAS400-2014.pdf

Your ANOVA is a General Linear Model (GLM)

and you can do Bayesian Inference in the General Linear Model using a Bayesian MCMC routine.

PROC ANOVA and PROC GLM do not have a BAYES statement, but PROC GENMOD does have a BAYES statement.

If needed you can use PROC GENMOD with a normal distribution and the identity link function (to mimic PROC GLM).

BR,

Koen

Super User

## Re: Priori hypothesis

And better post it at Statistic Forum .
@StatDave @SteveDenham could give you a hand.
Super User

## Re: Priori hypothesis

/*
Did you mean "Test for Homogeneity of Variance" ?
Check HOVTEST option:
*/
proc glm data=sashelp.heart;
class bp_status;
model weight=bp_status/solution;
means bp_status/hovtest;
quit;

## Re: Priori hypothesis

If it is as simple as comparing means, you can use a MEANS statement (ANOVA, GLM) or an LSMEANS statement (GLM and others). If the comparison is more complicated, use an LSMESTIMATE statement to construct the a priori comparison you wish to make.

SteveDenham

Discussion stats
• 4 replies
• 169 views
• 0 likes
• 4 in conversation