I’ve been playing around with a Stan version of the semilocal linear trend model , as found in the R package bsts . The model both automatically scales and matches bsts’s fit fairly well. Still, I’m hoping to get some feedback about: starting points mu[1] and nu[1], choice of priors (I offer my th…

Semilocal Linear Trend Model, opinions/suggestions welcome

andrewgelman July 16, 2018, 8:24pm 3

Hi, I don’t really have anything to add on this except to suggest simulating some fake data from the prior and see what these simulated time series look like.
Andrew

Topic		Replies	Views
Local linear trend model Modeling rstan , fitting-issues , state-space	7	2272	April 3, 2021
Fitting trend/bias in an AR(1) timeseries Modeling	2	1054	August 3, 2018
Bsts and forecasting Modeling	3	1659	July 18, 2019
Bayesian structural time series modeling Modeling fitting-issues	4	7896	October 19, 2017
Simultaneous modeling of long-term and short-term trends in a state-space model Modeling	3	347	March 4, 2024

Semilocal Linear Trend Model, opinions/suggestions welcome

Related topics