Root Finding Function in Stan

[Shoshana Vasserman]

Hi!

I need to do a fairly simple 1-dimensional convex optimization in every iteration of my estimation and so I need either a convex optimizer or a (numerical) root finder. I know that the Stan dev team is working on pushing a build with a full fledged function for this, but it seems like it'll be a while till it is pushed to the public build so I'm hoping to use a custom function in stan for myself.

I don't have much experience with writing root-finders - does anyone have an example or some guidance that might be of help? Anything would be greatly appreciated!

Thanks!

-Shosh

[Ben Goodrich]

On Monday, May 1, 2017 at 4:12:14 PM UTC-4, Shoshana Vasserman wrote:

I need to do a fairly simple 1-dimensional convex optimization in every iteration of my estimation and so I need either a convex optimizer or a (numerical) root finder. I know that the Stan dev team is working on pushing a build with a full fledged function for this, but it seems like it'll be a while till it is pushed to the public build so I'm hoping to use a custom function in stan for myself.

I don't have much experience with writing root-finders - does anyone have an example or some guidance that might be of help? Anything would be greatly appreciated!

For 1 dimension, use something from Boost

http://www.boost.org/doc/libs/1_63_0/libs/math/doc/html/math_toolkit/roots.html

with the mechanism for including C++ code

https://cran.r-project.org/web/packages/rstan/vignettes/external.html

Ben

[Shoshana Vasserman]

Thank you very much!

-Shosh

--
You received this message because you are subscribed to a topic in the Google Groups "Stan users mailing list" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/stan-users/Oh-JpUoRlAc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to stan-users+unsubscribe@googlegroups.com.
To post to this group, send email to stan-...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.