NEWS.md

In bayesforecast: Bayesian Time Series Modeling with Stan

bayesforecast NEWS

Bayesforecast 1.0.1. Date: 16/06/2021

Features:

• Local level models using local_level() and stan_local_level() methods

• Adding introductory vignette.

• Add mcmc_plot for posterior check

Improvements:

• adding seasonal logical function to auto.sarima() function for avoiding Seasonal models.

Changes:

• No current changes

Fixes:

• Correcting the default number of iterations per chain to 2000.

• Correcting ets Stan code, for seasonality.

• correcting bug in predictive_error() function.

-fix doi:10.2307/1392556 in GARCH vignette

-fix URl in GARCH vignette

Bayesforecast 0.1.1. Date: 29/01/2021

Features:

• bayesforecast package posted on CRAN

Improvements:

• No current changes

Changes:

• No current changes

Fixes:

• Correcting the default number of iterations per chain to 2000.

• Correcting ets stan code, for seasonality.

Bayesforecast 0.0.1. Date: 07/12/2020

Features:

• bayesforecast package created.

• Less stan models to compile than the previous varstan package.

• simplified interface.

Improvements:

• stan_model() functions created for similar syntax to brms and rstanarm packages.

• forecast EST methods implemented for Bayesian estimation.

• local global trend methods implemented from the Orbit package.

• automatic forecast methods for similar syntax to forecast package.

Changes:

• No current changes

Fixes:

• No current changes

bayesforecast documentation built on June 17, 2021, 5:14 p.m.