It's a development version. So I strong recommend you to update whenever you use it. You can install rafa from github with:
library(devtools)
devtools::install_github("leripio/rafa")
library(rafa)
library(rafa)
my_fc <- auto_forecast(USAccDeaths, test = c(1976,12))
my_fc$fc
This project is intended to develop an automatic forecasting algorithm in R language with robust specification and confidence intervals calculations. The main goal of this algorithm is to provide the best possible forecast for a time series based on univariate models to serve either as a benchmark for variables which we have a reference model or the reference itself when we do not have a reference model.
More specifically, it relies heavily on the R forecast package by Rob Hyndman and it is divided in 4 steps:
I intend to incorporate the "rectify" strategy proposed by Hyndman & Taieb (2012) so as the prediction errors can be used to further improve forecasts' accuracy.
This tool has been developed at Codeplan as an effort to obtain reliable and timely forecasts from economic variables. However, it is important to highlight that it serves only as a guide to economic analysis and should not be taken as an official tool neither from Codeplan nor from Distrito Federal Government.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.