data_prep_single_ts: A function to prepare a single time series for forecasting...

In vidarsumo/sumots: Helper functions for Sumo Analytics

Description

data_prep_single_ts() is function that allows you to create machine ready data

Usage

data_prep_single_ts(
  data,
  dep_var,
  date_group = "day",
  clean_outliers = TRUE,
  drop_na = TRUE,
  prepros = "log",
  spline_degree = 3,
  fourier_k = 5,
  horizon = 30,
  use_xreg = FALSE,
  zero_out_fridagar = today() + years(1),
  fridagar = FALSE
)

Arguments

data	A data frame that contains only two variables, date and the outcome
dep_var	Unquoted name of the dependent variable
date_group	Frequency of the data, defaults to day
drop_na	When creating data_prepared_tbl, should NA's be dropped. Defaults to TRUE
prepros	The preprocessing of the outcome variable. Defaults to log
spline_degree	The degree of the natural spline, defaults to 3
fourier_k	Number or orders to include for each sine/cosine
use_xreg	Are you using external regressors, defaults to FALSE
zero_out_fridagar	Should holidays be set to zero for the forecast horizon due to e.g. covid, defaults to 1 year into the future
fridagar	Do you want to use holidays, defaults to FALSE
horzon	The forecasting horizon, defaults to 30

vidarsumo/sumots documentation built on June 29, 2021, 4:23 a.m.