cforecastIn: Evaluate autoregressive moving average model (ARMA) model...

In snarf-snarf/karma: Complementary tools for automated time-series machine learning.

Description

Evaluate autoregressive moving average model (ARMA) model accuracy on in-sample data.

Usage

cforecastIn(arima_obj, yt, constant = "auto")

Arguments

arima_obj	input model object (output of arma() or Arima())
yt	input time series vector
terms	input AR, MA terms list
lib	input library name string used to create model object ('tseries' or 'forecast'). Defaults to 'tseries'.

Value

Forecasted time series vector.

See Also

tseries, forecast

Examples

library(tseries)
#Fit ARMA model:
dyt = diff(WWWusage)
arma1 = arma( x = dyt, lag = list(ar = c(1,2,3) ))
yt_hat = forecastIn( dyt, arma1, terms=list(ar = c(1,2,3)) )
#Plot predicted series:
plot(yt_hat, col="red", type="l") 
points(dyt, type="l", col="black")

snarf-snarf/karma documentation built on May 24, 2019, 7:19 a.m.