out_of_sample.listmtar: Computing Out-of-Sample predictive accuracy measures

In mtarm: Bayesian Estimation of Multivariate Threshold Autoregressive Models

Computing Out-of-Sample predictive accuracy measures

Description

Computes Out-of-Sample predictive accuracy measures for an object of class listmtar.

Usage

## S3 method for class 'listmtar'
out_of_sample(
  x,
  newdata,
  n.ahead = NULL,
  credible = 0.95,
  FUN = mean,
  rolling = NULL,
  ...
)

Arguments

x	An object of class listmtar returned by the routine mtar_grid().
newdata	A data.frame containing future values of the threshold series (if included in the fitted model), the exogenous series (if included in the fitted model), and the realized values of the output series.
n.ahead	A positive integer specifying the number of steps ahead to forecast.
credible	An optional numeric value in (0,1) specifying the level of the required prediction intervals. By default, credible is set to 0.95.
FUN	An optional function used to summarize the n.ahead values computed for each predictive accuracy measure. By default, FUN is set to mean.
rolling	An optional positive integer specifying the rolling-window size used for forecasting. By default, rolling = NULL, indicating that rolling-window forecasting is not performed.
...	optional arguments to FUN.

Examples

###### Example 1: Returns of the closing prices of three financial indexes
data(returns)
fit1 <- mtar_grid(~ COLCAP + BOVESPA | SP500, data=returns, row.names=Date,
                  subset={Date<="2015-12-07"}, dist=c("Gaussian","Student-t",
                  "Slash","Laplace"), nregim.min=2, nregim.max=3, p.min=2,
                  p.max=2, n.burnin=1000, n.sim=2000, n.thin=2,
                  plan_strategy="multisession")
oos1 <- out_of_sample(fit1, newdata=subset(returns, Date>"2015-12-07"),
                      n.ahead=75, FUN=median)
oos1

###### Example 2: Rainfall and two river flows in Colombia
data(riverflows)
fit2 <- mtar_grid(~ Bedon + LaPlata | Rainfall, data=riverflows,
                  row.names=Date, subset={Date<="2009-02-13"},dist="Laplace",
                  nregim.min=2, nregim.max=3, p.min=1, p.max=3,n.burnin=1000,
                  n.sim=2000, n.thin=2, plan_strategy="multisession")
oos2 <- out_of_sample(fit2, newdata=subset(riverflows, Date>"2009-02-13"),
                      n.ahead=60, FUN=median)
oos2

###### Example 3: Temperature, precipitation, and two river flows in Iceland
data(iceland.rf)
fit3 <- mtar_grid(~ Jokulsa + Vatnsdalsa | Temperature | Precipitation,
                  data=iceland.rf,subset={Date<="1974-11-06"},row.names=Date,
                  dist=c("Slash","Student-t"), nregim.min=1, nregim.max=2,
                  p.min=15, p.max=15, q.min=4, q.max=4, d.min=2, d.max=2,
                  n.burnin=1000, n.sim=2000, n.thin=2,
                  plan_strategy="multisession")
oos3 <- out_of_sample(fit3, newdata=subset(iceland.rf, Date>"1974-11-06"),
                      n.ahead=55, FUN=median)
oos3

###### Example 4: U.S. stock returns
data(US.returns)
fit4 <- mtar_grid(~ CCR | dVIX, data=US.returns, subset={Date<="2025-11-28"},
                  row.names=Date, dist=c("Laplace","Student-t","Slash"),
                  nregim.min=2, nregim.max=2, p.min=3, p.max=3, d.min=3,
                  d.max=3, n.burnin=1000, n.sim=2000, n.thin=2,
                  plan_strategy="multisession")
oos4 <- out_of_sample(fit4, newdata=subset(US.returns, Date>"2025-11-28"),
                      n.ahead=100, FUN=median)
oos4

mtarm documentation built on June 12, 2026, 5:07 p.m.