joinpoint regression using R-INLA

Rudy B

unread,

Jun 12, 2020, 11:20:11 AM6/12/20

to R-inla discussion group

Hi, Has anyone used joinpoint regression (e.g. coefficients vary by time periods) within the Bayes setting? Regards, Rudy

INLA help

unread,

Jun 12, 2020, 1:26:27 PM6/12/20

to R-inla discussion group, Rudy B

Varying regression is done using a model for the varying part, and then add the covariate, like

f(time, z, model=“rw2”, constr=FALSE)

--

Håvard Rue

Helpdesk

he...@r-inla.org

On 12 Jun 2020, 18:20 +0300, Rudy B <aniruddha...@gmail.com>, wrote:

Hi, Has anyone used joinpoint regression (e.g. coefficients vary by time periods) within the Bayes setting? Regards, Rudy

--
You received this message because you are subscribed to the Google Groups "R-inla discussion group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to r-inla-discussion...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/r-inla-discussion-group/019665c3-42bb-42fa-9a83-248eb531e9efo%40googlegroups.com.

Aniruddha Banerjee

unread,

Jun 12, 2020, 1:48:08 PM6/12/20

to INLA help, R-inla discussion group

Thanks!

--

Aniruddha Banerjee, Ph.D.

Jonathan Simkin

unread,

Jul 5, 2022, 1:39:37 PM7/5/22

to R-inla discussion group

Hey - I'm interested in replicating JoinPoint type analysis in INLA as well. This can be done with the mcp R package (https://lindeloev.github.io/mcp/), not sure how to do it here.

From my understanding, joinpoint and mcp are essentially finding breakpoints and estimating the slope between the segments.. and you can specify how many breakpoints to segment the linear trend by (e.g. 1, 2, 3.. n breakpoints). Is there a way to do that here in INLA?