archtest |

Test for autoregressive conditional heteroskedasticity (ARCH).

Carries out Lagrange Multiplier (LM) tests for ARCH in the residuals of a single least squares equation.

Syntax

eq_name.archtest(options)

Options

You must specify the order of ARCH for which you wish to test. The number of lags to be included in the test equation should be provided in parentheses after the arch keyword.

Other Options:

prompt | Force the dialog to appear from within a program. |

p | Print output from the test. |

Examples

equation eq1.ls output c labor capital

eq1.archtest(4)

Regresses OUTPUT on a constant, LABOR, and CAPITAL, and tests for ARCH up to order 4.

equation eq1.arch sp500 c

eq1.archtest(4)

Estimates a GARCH(1,1) model with mean equation of SP500 on a constant and tests for additional ARCH up to order 4. Note that when performing an archtest as a view off of an estimated arch equation, EViews will use the standardized residuals (the residual of the mean equation divided by the estimated conditional standard deviation) to form the test.

Cross-references

See “ARCH LM Test” for further discussion of testing ARCH and “ARCH and GARCH Estimation” for a general discussion of working with ARCH models in EViews.

See also Equation::hettest for a more full-featured version of this test.