Scale forecast analysis with a Modeltime Table
Designed to perform forecasts at scale using models created with
modeltime, parsnip, workflows, and regression modeling extensions
in the tidymodels ecosystem.
modeltime_table(...) as_modeltime_table(.l)
... |
Fitted |
.l |
A list containing fitted |
modeltime_table():
Creates a table of models
Validates that all objects are models (parsnip or workflows objects) and all models have been fitted (trained)
Provides an ID and Description of the models
as_modeltime_table():
Converts a list of models to a modeltime table. Useful if programatically creating
Modeltime Tables from models stored in a list.
library(tidyverse)
library(lubridate)
library(timetk)
library(parsnip)
library(rsample)
# Data
m750 <- m4_monthly %>% filter(id == "M750")
# Split Data 80/20
splits <- initial_time_split(m750, prop = 0.9)
# --- MODELS ---
# Model 1: auto_arima ----
model_fit_arima <- arima_reg() %>%
set_engine(engine = "auto_arima") %>%
fit(value ~ date, data = training(splits))
# ---- MODELTIME TABLE ----
# Make a Modeltime Table
models_tbl <- modeltime_table(
model_fit_arima
)
# Can also convert a list of models
list(model_fit_arima) %>%
as_modeltime_table()
# ---- CALIBRATE ----
calibration_tbl <- models_tbl %>%
modeltime_calibrate(new_data = testing(splits))
# ---- ACCURACY ----
calibration_tbl %>%
modeltime_accuracy()
# ---- FORECAST ----
calibration_tbl %>%
modeltime_forecast(
new_data = testing(splits),
actual_data = m750
)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.