If the root of the characteristic equation for a time series is one then that series is said to have a unit root. Such series are nonstationary. g13awf returns one of three types of (augmented) Dickey–Fuller test statistic: , or , used to test for a unit root, a unit root with drift or a unit root with drift and a deterministic time trend, respectively.
To test whether a time series,
, for
, has a unit root, the regression model
is fitted and the test statistic
constructed as
where
is the difference operator, with
, and where
and
are the least squares estimate and associated standard error for
respectively.
To test for a unit root with drift the regression model
is fit and the test statistic
constructed as
To test for a unit root with drift and deterministic time trend the regression model
is fit and the test statistic
constructed as
The distributions of the three test statistics;
,
and
, are nonstandard. An associated probability can be obtained from
g01ewf.
Dickey A D (1976) Estimation and hypothesis testing in nonstationary time series PhD Thesis Iowa State University, Ames, Iowa
Dickey A D and Fuller W A (1979) Distribution of the estimators for autoregressive time series with a unit root J. Am. Stat. Assoc. 74 366 427–431
If on entry
or
, explanatory error messages are output on the current error message unit (as defined by
x04aaf).
None.
None.
In this example a Dickey–Fuller unit root test is applied to a time series related to the rate of the earth's rotation about its polar axis.