Joint F test on interaction effects

[Samira Choudhury]

Hi everyone,
For one of my papers, I am using Oaxaca decomposition method looking at caste differences in consumption. To see whether there is evidence of differences in the effects of the determinants, we have to first run a regression of household Fruit and veg consumption (dependent variable) on the covariates interacted with the dummy variable separating low castes from high castes.

What is the STATA code for doing the joint F test on these interaction effects after running the above regression?

Many thanks for your help,
Samira.

[Carlo Lazzaro]

Samira:
do you mean somethuing along the following lines?

. sysuse auto.dta
(1978 Automobile Data)

. regress price i.foreign##i.rep78
note: 1.foreign#1b.rep78 identifies no observations in the sample
note: 1.foreign#2.rep78 identifies no observations in the sample
note: 1.foreign#5.rep78 omitted because of collinearity

      Source |       SS           df       MS      Number of obs   =        69
-------------+----------------------------------   F(7, 61)        =      0.39
       Model |    24684607         7  3526372.43   Prob > F        =    0.9049
    Residual |   552112352        61  9051022.16   R-squared       =    0.0428
-------------+----------------------------------   Adj R-squared   =   -0.0670
       Total |   576796959        68  8482308.22   Root MSE        =    3008.5

-------------------------------------------------------------------------------
        price |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
      foreign |
     Foreign  |   2088.167   2351.846     0.89   0.378     -2614.64    6790.974
              |
        rep78 |
           2  |   1403.125   2378.422     0.59   0.557    -3352.823    6159.073
           3  |   2042.574   2204.707     0.93   0.358    -2366.011    6451.159
           4  |   1317.056   2351.846     0.56   0.578    -3385.751    6019.863
           5  |       -360   3008.492    -0.12   0.905    -6375.851    5655.851
              |
foreign#rep78 |
   Foreign#1  |          0  (empty)
   Foreign#2  |          0  (empty)
   Foreign#3  |  -3866.574   2980.505    -1.30   0.199    -9826.462    2093.314
   Foreign#4  |  -1708.278   2746.365    -0.62   0.536    -7199.973    3783.418
   Foreign#5  |          0  (omitted)
              |
        _cons |     4564.5   2127.325     2.15   0.036      310.651    8818.349
-------------------------------------------------------------------------------

. mat list e(b)

e(b)[1,18]
             0b.           1.          1b.           2.           3.           4.           5.
        foreign      foreign        rep78        rep78        rep78        rep78        rep78
y1            0    2088.1667            0     1403.125    2042.5741    1317.0556         -360

     0b.foreign#  0b.foreign#  0b.foreign#  0b.foreign#  0b.foreign#  1o.foreign#  1o.foreign#
       1b.rep78     2o.rep78     3o.rep78     4o.rep78     5o.rep78     1b.rep78     2o.rep78
y1            0            0            0            0            0            0            0

      1.foreign#   1.foreign#  1o.foreign#            
        3.rep78      4.rep78     5o.rep78        _cons
y1   -3866.5741   -1708.2778            0       4564.5

. test 1o.foreign#4.rep78=1.foreign#3.rep78

 ( 1)  - 1.foreign#3.rep78 + 1.foreign#4.rep78 = 0

       F(  1,    61) =    0.87
            Prob > F =    0.3550

.

[Samira Choudhury]

Hi Carlo,
Thanks for your response. I don't think I explained it quite clearly. I am running the following regression where I am interacting all the covariates with the low vs high castes (scheduledtribecaste).


These are my covariates-
logmpce logoutlierrelativeprice hh_size c0_5 educfsp - continuous variables
femhh rural agri hindu state - dummy variables

reg aeufruitvegg logmpce logoutlierrelativeprice hh_size c0_5 educfsp femhh rural agri hindu i.state i.scheduledtribecaste i.scheduledtribecaste#c.(logmpce logoutlierrelativeprice hh_size c0_5 educfsp) /*
*/ i.scheduledtribecaste#i.(femhh rural agri hindu state) [aw=hhwt], vce(cluster FSU_Serial_No)

Then do a Joint F statistics for ALL interactions including the dummy on its own.

test 1.scheduledtribecaste 1.scheduledtribecaste#c.logmpce 1.scheduledtribecaste#c.logoutlierrelativeprice 1.scheduledtribecaste#c.hh_size /*
*/ 1.scheduledtribecaste#c.c0_5 1.scheduledtribecaste#c.educfsp 1.scheduledtribecaste#i.femhh 1.scheduledtribecaste#i.rural 1.scheduledtribecaste#i.agri 1.scheduledtribecaste#i.hindu 1.scheduledtribecaste#i.state

But now I'm getting this error "i: operator invalid".

How do I solve this problem?

Thanks a lot for your help,
Samira.

[Carlo Lazzaro]

Samira:
you should take the names of the coefficients from:

mat list e(b)

.

Stata assigns those name for calculation purposes and they should be used with postestimation commands such as -test-.

[Samira Choudhury]

Thanks Carlo.

[Samira Choudhury]

I still get the same error when running test command.

test ..... 1.scheduledtribecaste#i.rural

rural is a dummy variable.


Thanks for your help,
Samira.

[Carlo Lazzaro]

Samira:
in the following example I cannot reproduce your problem:

use "C:\Program Files (x86)\Stata15\ado\base\a\auto.dta"
. reg price i.foreign##i.rep78
note: 1.foreign#1b.rep78 identifies no observations in the sample
note: 1.foreign#2.rep78 identifies no observations in the sample
note: 1.foreign#5.rep78 omitted because of collinearity

      Source |       SS           df       MS      Number of obs   =        69
-------------+----------------------------------   F(7, 61)        =      0.39
       Model |    24684607         7  3526372.43   Prob > F        =    0.9049
    Residual |   552112352        61  9051022.16   R-squared       =    0.0428
-------------+----------------------------------   Adj R-squared   =   -0.0670
       Total |   576796959        68  8482308.22   Root MSE        =    3008.5

-------------------------------------------------------------------------------
        price |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
      foreign |
     Foreign  |   2088.167   2351.846     0.89   0.378     -2614.64    6790.974
              |
        rep78 |
           2  |   1403.125   2378.422     0.59   0.557    -3352.823    6159.073
           3  |   2042.574   2204.707     0.93   0.358    -2366.011    6451.159
           4  |   1317.056   2351.846     0.56   0.578    -3385.751    6019.863
           5  |       -360   3008.492    -0.12   0.905    -6375.851    5655.851
              |
foreign#rep78 |
   Foreign#1  |          0  (empty)
   Foreign#2  |          0  (empty)
   Foreign#3  |  -3866.574   2980.505    -1.30   0.199    -9826.462    2093.314
   Foreign#4  |  -1708.278   2746.365    -0.62   0.536    -7199.973    3783.418
   Foreign#5  |          0  (omitted)
              |
        _cons |     4564.5   2127.325     2.15   0.036      310.651    8818.349
-------------------------------------------------------------------------------

. mat list e(b)

e(b)[1,18]
             0b.           1.          1b.           2.           3.           4.           5.
        foreign      foreign        rep78        rep78        rep78        rep78        rep78
y1            0    2088.1667            0     1403.125    2042.5741    1317.0556         -360

     0b.foreign#  0b.foreign#  0b.foreign#  0b.foreign#  0b.foreign#  1o.foreign#  1o.foreign#
       1b.rep78     2o.rep78     3o.rep78     4o.rep78     5o.rep78     1b.rep78     2o.rep78
y1            0            0            0            0            0            0            0

      1.foreign#   1.foreign#  1o.foreign#            
        3.rep78      4.rep78     5o.rep78        _cons
y1   -3866.5741   -1708.2778            0       4564.5

. test 1.foreign 1b.rep78 2.rep78 3.rep78 4.rep78 5.rep78 0b.foreign#1b.rep78 0b.foreign#2o.rep78 0b.foreign#3o.rep78 0b.foreign#4o.rep78 0b.foreign#5o.rep78 1o.foreign#1b.rep78 1o.foreign#2o.rep78 1.foreign#3.rep78 1.foreign#4.rep78 1o.foreign#5o.rep78

 ( 1)  1.foreign = 0
 ( 2)  1b.rep78 = 0
 ( 3)  2.rep78 = 0
 ( 4)  3.rep78 = 0
 ( 5)  4.rep78 = 0
 ( 6)  5.rep78 = 0
 ( 7)  0b.foreign#1b.rep78 = 0
 ( 8)  0b.foreign#2o.rep78 = 0
 ( 9)  0b.foreign#3o.rep78 = 0
 (10)  0b.foreign#4o.rep78 = 0
 (11)  0b.foreign#5o.rep78 = 0
 (12)  1o.foreign#1b.rep78 = 0
 (13)  1o.foreign#2o.rep78 = 0
 (14)  1.foreign#3.rep78 = 0
 (15)  1.foreign#4.rep78 = 0
 (16)  1o.foreign#5o.rep78 = 0
       Constraint 2 dropped
       Constraint 7 dropped
       Constraint 8 dropped
       Constraint 9 dropped
       Constraint 10 dropped
       Constraint 11 dropped
       Constraint 12 dropped
       Constraint 13 dropped
       Constraint 16 dropped

       F(  7,    61) =    0.39
            Prob > F =    0.9049

.

You may want to add one coefficient at time and see when -test- starts gasping.

[Samira Choudhury]

okay thanks a lot Carlo!