Reset test

[sony vebiyanto]

hi,,I'm new to using STATA I have a problem when I want to do a RESET test on a gravity model regression with PPML
I used the following command:
ppml ekspor PAIR_FE* year_* ln_yi ln_yj ln_rem_exp ln_rem_imp fta1 fta2 fta3
predict fit, xb
gen fit2=fit^2
test fit2=0

But the result shows that :
fit2 is not found
r(111);

can someone explain what i missed or what i did wrong? fyi, I used 3 dummy variables of FTA and the best effect are using country-pair and year effect
Thx

[Jared Greathouse]

Please see FAQ Section 12 on how to ask a question.

[Carlo Lazzaro]

Sony:
welocme to this forum.
As an aside to Jared's wise recommendation, you cannot invoke a -posyestimation- commnd, such as -test-, without a previous estimate.
See the following toy-example on -regress-:

. use "C:\Program Files\Stata17\ado\base\a\auto.dta"
(1978 automobile data)

. regress price i.foreign

      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(1, 72)        =      0.17
       Model |  1507382.66         1  1507382.66   Prob > F        =    0.6802
    Residual |   633558013        72  8799416.85   R-squared       =    0.0024
-------------+----------------------------------   Adj R-squared   =   -0.0115
       Total |   635065396        73  8699525.97   Root MSE        =    2966.4

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
     foreign |
    Foreign  |   312.2587   754.4488     0.41   0.680    -1191.708    1816.225
       _cons |   6072.423    411.363    14.76   0.000     5252.386     6892.46
------------------------------------------------------------------------------

. predict fitted, xb

. g sq_fitted=fitted^2

. test sq_fitted=0
sq_fitted not found
r(111);

. regress price fitted sq_fitted
note: fitted omitted because of collinearity.

      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(1, 72)        =      0.17
       Model |  1507382.66         1  1507382.66   Prob > F        =    0.6802
    Residual |   633558013        72  8799416.85   R-squared       =    0.0024
-------------+----------------------------------   Adj R-squared   =   -0.0115
       Total |   635065396        73  8699525.97   Root MSE        =    2966.4

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
      fitted |          0  (omitted)
   sq_fitted |   .0000803    .000194     0.41   0.680    -.0003064    .0004669
       _cons |    3112.32   7384.262     0.42   0.675    -11607.94    17832.58
------------------------------------------------------------------------------

. test sq_fitted=0

 ( 1)  sq_fitted = 0

       F(  1,    72) =    0.17
            Prob > F =    0.6802

.

[sony vebiyanto]

Mr. Carlo Lazarro
Thx for your response

It works,, although the results are not as desired because they did not pass the Ramsey test

[Carlo Lazzaro]

Sony:
check whether your model has all the predictors and/or interactions plugged in.