Open kuchaale opened 2 years ago
Hi Ales,
Thanks for this — looks very cool, do you want to tell me exactly how F and G need modifying and I’ll look into doing it?
Thanks!
On 29 Jun 2022, at 12:43, Ales Kuchar @.***> wrote:
Dear @justinalsing https://github.com/justinalsing, I use a combination of dynamical linear models with shrinkage priors (Arnold, 2015) http://docs.jrnold.me.s3.amazonaws.com/dlm-shrinkage-2015-10-14.pdf or repo here https://github.com/jrnold/dlm-shrinkage. The approach has several advantages:
not required to specify a number of change points (discontinuities) in advance, it can be estimated from the data flexible (non-parametric) model for non-linear backround trend similarly to dlmmc https://github.com/justinalsing/dlmmc However, I would like to use it in combination with seasonality and external regressors as in the case of dlmmc https://github.com/justinalsing/dlmmc. Tests on temperature time-series with discontinuities (see below) https://user-images.githubusercontent.com/6815953/176417032-dacc746a-0199-440d-9463-0c071164938c.png showed that both packages reach similar temperature trends (compare below trends using dlm-shrinkage https://github.com/jrnold/dlm-shrinkage and dlmmc https://github.com/justinalsing/dlmmc). https://user-images.githubusercontent.com/6815953/176416819-3deaf511-dd42-4a3a-b1a4-336e2b816a64.png Nevertheless, I am not much proficient in Stan. Would you be interested in helping me to modify F and G matrices in dlmmc https://github.com/justinalsing/dlmmc to include shrinkage priors?
Thanks for your response Ales
— Reply to this email directly, view it on GitHub https://github.com/justinalsing/dlmmc/issues/9, or unsubscribe https://github.com/notifications/unsubscribe-auth/AF4N3MLC2BQVBDLTCM5CE73VRQSFVANCNFSM52FIGXIA. You are receiving this because you were mentioned.
Actually, more may be needed to modify. See e.g. poly_horseshoe.stan.mustache where all input parameters and model are introduced.
Dear @justinalsing, I use a combination of dynamical linear models with shrinkage priors (Arnold, 2015) or repo here. The approach has several advantages:
However, I would like to use it in combination with seasonality and external regressors as in the case of dlmmc. Tests on temperature time-series with discontinuities (see below)
showed that both packages reach similar temperature trends (compare below trends using dlm-shrinkage and dlmmc).
![image](https://user-images.githubusercontent.com/6815953/176416819-3deaf511-dd42-4a3a-b1a4-336e2b816a64.png)
Nevertheless, I am not much proficient in Stan. Would you be interested in helping me to modify
F
andG
matrices in dlmmc to include shrinkage priors?Thanks for your response Ales