0
$\begingroup$
Dickey-Fuller test for unit root                   Number of obs   =        21

                               ---------- Interpolated Dickey-Fuller ---------
                  Test         1% Critical       5% Critical      10% Critical
               Statistic           Value             Value             Value
------------------------------------------------------------------------------
 Z(t)            -14.272            -4.380            -3.600            -3.240
------------------------------------------------------------------------------
MacKinnon approximate p-value for Z(t) = 0.0000

------------------------------------------------------------------------------
D.TotalS     |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      TotalS |
         L1. |  -.9774459   .0684873   -14.27   0.000    -1.121332   -.8335593
      _trend |   .4713369   .0326362    14.44   0.000     .4027708     .539903
       _cons |   48.50135   3.370879    14.39   0.000     41.41939     55.5833
------------------------------------------------------------------------------

Specifically: Is it okay to have a p value of zero or one. In this example I have a p-value of zero. Also in a different case I am having a p value = 1.0000. Is that alright?

$\endgroup$
2
  • $\begingroup$ Could you elaborate on your question? What is it that you want to know exactly? (Presumably you are familiar with what the Number of obs is...) $\endgroup$ Feb 23, 2014 at 22:31
  • 1
    $\begingroup$ I am conducting unit root test for the first time. Is it okay to have a p value =O. Also in a different case I am having a p value = 1.0000. Is that alright? $\endgroup$
    – user40767
    Feb 23, 2014 at 22:35

1 Answer 1

1
$\begingroup$

The help for the dfuller command states:

"The null hypothesis is that the variable contains a unit root, [...]"

So obviously your null hypothesis can soundly be rejected. This is evidence that the underlying process is stationar. This is probably what you would like to see when your are about to infer something from your sample.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.