forecast.marx: Forecasting function for the MARX model

In MARX: Simulation, Estimation, Model Selection and Forecasting for MARX Models

Description

This function allows you to forecast with the mixed causal-noncausal model with possibly exogenous regressors.

Usage

forecast.marx(y, X, p_C, p_NC, X.for, h, M, N)

Arguments

y	Data vector y.
X	(optional) Matrix with data (column represent a series).
p_C	Number of lags (causal order).
p_NC	Number of leads (noncausal order).
X.for	(optional) Matrix with forecasted values for X (column represents series).
h	Forecast horizon h.
M	(optional) Truncation value M for MA representation. Default value: 50.
N	(optional) Number of simulations to forecast noncausal component. Default: 10,000.

Value

y.for

Vector containing forecasted values for y.

Author(s)

Sean Telg

Examples

## Forecasting MAR(0,1) model 4-periods ahead for lnbev (from dataset)
data <- MARX::dataset[,2]
y.for <- forecast.marx(y=data, p_C=0, p_NC=1, h=4, M=50, N=1000)

Example output

There were 17 warnings (use warnings() to see them)

MARX documentation built on May 2, 2019, 3:42 a.m.