Where are the critical values of DFGLS unit root test stored?

[Nora Kaimann]

A DFGLS test provides a result table with critical values at 1%, 5%, 10%, e.g. like this:

. dfgls ln_GDP_AUT_sa_shw1 if month >= 0 & month <= 156, maxlag(12)
 
DF-GLS for ln_CPI_AUT_sa_~1                              Number of obs =   144
 
               DF-GLS tau      1% Critical       5% Critical      10% Critical
  [lags]     Test Statistic        Value             Value             Value
------------------------------------------------------------------------------
    12           -1.373           -3.512            -2.809            -2.532
    11           -0.688           -3.512            -2.826            -2.548
    10           -0.672           -3.512            -2.843            -2.563
    9            -1.200           -3.512            -2.859            -2.578
    8            -1.457           -3.512            -2.875            -2.593
    7            -2.146           -3.512            -2.890            -2.607
    6            -1.849           -3.512            -2.904            -2.620
    5            -2.451           -3.512            -2.918            -2.632
    4            -2.090           -3.512            -2.930            -2.644
    3            -2.786           -3.512            -2.942            -2.655
    2            -3.344           -3.512            -2.954            -2.665
    1            -3.716           -3.512            -2.964            -2.674
 
Opt Lag (Ng-Perron seq t) = 12 with RMSE  .0070382
Min SC   =  -9.46413 at lag 12 with RMSE  .0070382
Min MAIC = -9.684341 at lag 12 with RMSE  .0070382 

But these critical values are not stored, e.g.

matrix list r(results) 

yields only the following table:

             k        MAIC         SIC        RMSE       DFGLS
r1          12  -9.6843409  -9.4641303   .00703824  -1.3725487
r1          11   -9.575829  -9.3296057   .00765897  -.68797794
r1          10  -9.5905321  -9.3638727   .00765991  -.67219615
r1           9   -9.501018  -9.3241323   .00794964  -1.1998016
r1           8  -9.4865895  -9.3479864   .00799211  -1.4566636
r1           7   -9.377198  -9.3221978   .00823675  -2.1459997
r1           6  -9.4126014  -9.3408127   .00830248  -1.8488197
r1           5  -9.3416657   -9.344457   .00843162   -2.450565
r1           4  -9.3767739  -9.3561343   .00852844  -2.0897924
r1           3  -9.2949919  -9.3554019   .00868007  -2.7859181
r1           2  -9.2513679  -9.3781925   .00873109  -3.3436531
r1           1  -9.2487849  -9.4117624   .00873521  -3.7159741 

Can anybody please tell me where to receive the 5% and 10% critical values, in order to write them with "putexcel" to an Excel file?

Thanks a lot!
Nora

[Sebastian Kripfganz]

Apparently, Stata does not return these critical values as stored results. You could compute them manually and create your own table or matrix based on equation (4) and the corresponding coefficients in Table 1 of Cheung and Lai (1995): Lag Order and Critical Values of a Modified Dickey-Fuller Test. Oxford Bulletin of Economics and Statistics 57 (3), 411-419.