autoreg_forc: Autoregression forecast

In lmForc: Linear Model Forecasting

Description

autoreg_forc takes a vector of realized values, an integer number of periods ahead to forecast, an integer number of lags to include in the autoregressive model, a period to end the initial model estimation and begin forecasting, an optional vector of time data associated with the realized values, and an optional integer number of past periods to estimate the model over. An AR(ar_lags) autoregressive model is originally estimated with realized values up to estimation_end minus the number of periods specified in estimation_window. If estimation_window is left NULL then the autoregressive model is estimated with all realized values up to estimation_end. The AR(ar_lags) model is estimated by regressing the realized values on the same realized values that have been lagged by one to ar_lags steps. The AR coefficients of this model are multiplied by lagged values and the present period realized value to create a forecast for one period ahead. If h_ahead is greater than one, this process of forecasting one period ahead is iteratively repeated so that the two period ahead forecast conditions on the one period ahead forecasted value and so on until a h_ahead forecast is obtained. This forecasting process is repeated for each period after estimation_end with AR model coefficients updating as more information would have become available to the forecaster. Optionally returns the coefficients used to create each forecast. Returns an autoregression forecast based on information that would have been available at the forecast origin and replicates the forecasts that an AR model would have produced in real-time.

Usage

autoreg_forc(
  realized_vec,
  h_ahead,
  ar_lags,
  estimation_end,
  time_vec = NULL,
  estimation_window = NULL,
  return_betas = FALSE
)

Arguments

realized_vec	Vector of realized values. This is the series that is being forecasted.
h_ahead	Integer representing the number of periods ahead that is being forecasted.
ar_lags	Integer representing the number of lags included in the AR model.
estimation_end	Value of any class representing when to end the initial coefficient estimation period and begin forecasting.
time_vec	Vector of any class that is equal in length to the realized_vec vector.
estimation_window	Integer representing the number of past periods that the autoregressive model should be estimated over in each period.
return_betas	Boolean, selects whether the coefficients used in each period to create the forecast are returned. If TRUE, a data frame of betas is returned to the Global Environment.

Value

Forecast object that contains the autoregression forecast.

See Also

For a detailed example see the help vignette: vignette("lmForc", package = "lmForc")

Examples

date <- as.Date(c("2010-03-31", "2010-06-30", "2010-09-30", "2010-12-31",
                  "2011-03-31", "2011-06-30", "2011-09-30", "2011-12-31",
                  "2012-03-31", "2012-06-30", "2012-09-30", "2012-12-31"))
y <- c(1.09, 1.71, 1.09, 2.46, 1.78, 1.35, 2.89, 2.11, 2.97, 0.99, 1.31, 2.33)
data <- data.frame(date, y)

autoreg_forc(
  realized_vec = data$y,
  h_ahead = 1L,
  ar_lags = 2L,
  estimation_end = as.Date("2011-06-30"),
  time_vec = data$date,
  estimation_window = 4L,
  return_betas = FALSE
)

autoreg_forc(
  realized_vec = data$y,
  h_ahead = 4L,
  ar_lags = 2L,
  estimation_end = 4L,
  time_vec = NULL,
  estimation_window = NULL
)

lmForc documentation built on Jan. 4, 2022, 1:11 a.m.