Dickey-Fuller F statistics for testing joint hypotheses

[Cheuk Wailun]

The command dfuller provides the Dickey-Fuller critical values for the unit root test, but how do I run the Dickey-Fuller F tests in Stata?

Thanks!

[Cheuk Wailun]

Bump. It can't be that Stata doesn't have this functionality, can it?

[Marcos Almeida]

Welcome to the Stata Forum.

Unfortunately, your query is far away from my field. But a quick look at the Manual(http://www.stata.com/manuals13/xtxtunitroot.pdf) provided me with this:

Fisher-type tests [...]
The null hypothesis being tested by xtunitroot fisher is that all panels contain a unit root.

You may also wish to check example 6, on pages 16 and 17.

Hopefully that helps.


Best,


Marcos

[Emad Shehata]

This example maybe help you to claculate F-Test of Dickey-Fuller unit-root test

clear all
input y
 99.2
   99
  100
111.6
122.2
117.6
121.1
  136
154.2
153.6
158.5
140.6
136.2
  168
154.3
  149
165.5
end

gen Time =_n
tsset Time
reg d.y l.y
test _cons=0 , notest
test l.y=0 , accumulate

 gen Time =_n

. tsset Time
        time variable:  Time, 1 to 17
                delta:  1 unit

. reg d.y l.y

      Source |       SS       df       MS              Number of obs =      16
-------------+------------------------------           F(  1,    14) =    1.51
       Model |  236.129283     1  236.129283           Prob > F      =  0.2392
    Residual |  2187.16985    14  156.226418           R-squared     =  0.0974
-------------+------------------------------           Adj R-squared =  0.0330
       Total |  2423.29914    15  161.553276           Root MSE      =  12.499

------------------------------------------------------------------------------
         D.y |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
           y |
         L1. |  -.1731765   .1408612    -1.23   0.239    -.4752936    .1289406
             |
       _cons |   27.10154   18.93342     1.43   0.174    -13.50661    67.70969
------------------------------------------------------------------------------

. test _cons=0 ,  notest

 ( 1)  _cons = 0

. test l.y=0 , accumulate

 ( 1)  _cons = 0
 ( 2)  L.y = 0

       F(  2,    14) =    1.63
            Prob > F =    0.2301

[Emad Shehata]

t-test = -1.23
is identical to:

. dfuller y , lag(0)

Dickey-Fuller test for unit root                   Number of obs   =        16

                               ---------- Interpolated Dickey-Fuller ---------
                  Test         1% Critical       5% Critical      10% Critical
               Statistic           Value             Value             Value
------------------------------------------------------------------------------
 Z(t)             -1.229            -3.750            -3.000            -2.630
------------------------------------------------------------------------------
MacKinnon approximate p-value for Z(t) = 0.6608