get_ts_forecasts: Perform a suite of time-series only forecasts

In weecology/biodivcast:

Description

Performs time-series only forecasting using naive, average, and ARIMA models for each group in a grouped data frame of time-series.

Usage

get_ts_forecasts(grouped_tsdata, timecol, responsecol, exogcols, lag = 1,
  pred_int_levels = c(80, 95))

Arguments

grouped_tsdata	data.frame is a grouped data frame where the group indicates how to split the data frame for running time-series models on each group
timecol	str name of the column that includes the time variable
responsecol	str name of the column that includes the response variable
lag	numeric how many time-steps to forecast and also how many time-steps in the current time-series to ignore when fitting.

Value

data.frame

TODO: separate time-steps to hold out from fitting from time-steps to forecast, since when not hindcasting these may be different

weecology/biodivcast documentation built on May 20, 2019, 1:25 p.m.