This function takes in a time series data and generates a forecast for specified number of periods.
The function fits an ARIMA model to the data and uses it to generate the forecast.
If the data has a clear seasonal pattern, the function fits a seasonal ARIMA model.
The function returns two lists: obs_fitted (the fitted values over the observation period)
and pred_trend (the predicted trend for the forecast horizon).
Usage
forecast_time_series(ts_data, forecast_horizon, season_period = NULL)
Arguments
- ts_data
a time series object or a numeric vector
- forecast_horizon
a positive integer specifying the number of periods to forecast
- season_period
an optional parameter specifying the seasonal period of the time series
Value
a list containing the observations and predicted trend for the forecast horizon
Examples
# generate a time series data
ts_data <- ts(rnorm(100), frequency = 12, start = c(2014, 1))
# generate a forecast for next 6 periods using auto.arima() with default settings
forecast_time_series(ts_data, 6)
#> Error in forecast_time_series(ts_data, 6): object 'su_yaml' not found