com.imsl.test.example.stat

Class ARMAOutlierIdentificationEx3

java.lang.Object

com.imsl.test.example.stat.ARMAOutlierIdentificationEx3

public class ARMAOutlierIdentificationEx3
extends Object

Forecasts an $\text{ARMA}(2,1)$ time series contaminated by outliers.

This example is a realization of an $\text{ARMA}(2,1)$ process described by the model $Y_t-Y_{t-1}+0.24Y_{t-2} = 10.0+a_t+0.5a_{t-1},\, \{a_t\}$ a Gaussian White noise process. An additive outlier with $\omega_1=4.5$ was added at time point $t=150$, a temporary change outlier with $\omega_2=3.0$ was added at time point $t=200$.

Outliers were artificially added to the outlier free series $\{Y-t\}_{t=1,\ldots, 280}$ at time points $t=150$ (level shift with $\omega_1=+2.5$) and $t=200$ (additive outlier with $\omega_2=+3.2$), resulting in the outlier contaminated series $\{Z_t\}_{t=1,\ldots,280}$. For both series, forecasts were determined for time points $t=281,\ldots,290$ and compared with the actual values of the series.

See Also:

Code, Output

Constructor Summary

Constructors

Constructor and Description
ARMAOutlierIdentificationEx3()

Method Summary

Modifier and Type	Method and Description
static void	main(String[] args)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

ARMAOutlierIdentificationEx3

public ARMAOutlierIdentificationEx3()

Method Detail

main

public static void main(String[] args)
                 throws Exception

Throws:

Exception