R framework that makes evaluation of forecasts easy.
This is a very early prototype. Don't touch or you will get lost in time.
remotes::install_github("christophsax/timemachine")
library(timemachine)
Since version 0.0.2, 'history' and 'dates' are not set via options anymore, and must be provided explicitly.
Run timemachine
with one or several forecast models. Build the models as you
would build them in standard R. timemachine
exposes the data, evaluates the
forecasts and collects the data.
The expression must evaluate to a tsboxable object. E.g., a ts
, xts
, or
data.frame
object.
library(forecast)
timemachine(
{
m <- forecast(auto.arima(GDP.CH))
m$mean
},
history = swiss_history,
dates = seq(as.Date("1979-10-01"), to = as.Date("1979-12-01"), by = "month")
)
More advaned examples are on the help page of ?timemachine
.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.