Errors with nnest command in stata

[Muhammad Ibrahim Shah]

Dear all, I am running nnest command in stata to compare between two models. My commands are as follows:


reg lnprice lnlotsize beddummy1 beddummy2 beddummy3 airco bathrms
estimates store m1

reg lnprice lnlotsize bedrooms airco bathrms
estimates store m2

nnest m1 m2

Output: (variable m1 not found
r(111);

So I run the model stata generated

nnest _est_m1 _est_m2


Competing Models
--------------------------------------------------
M1 : Y = [lnprice]
X = [lnlotsize bedrooms airco bathrms]
M2 : Y = [lnprice]
Z = [_est_m1 _est_m2]
--------------------------------------------------
J test for non-nested models
--------------------------------------------------
Dist Stat P>Stat
H0:M1 / H1:M2 t(541) . .
H0:M2 / H1:M1 t(544) 26.71 0.000
--------------------------------------------------
Cox-Pesaran test for non-nested models
--------------------------------------------------
Dist Stat P>Stat
H0:M1 / H1:M2 N(0,1) . .
H0:M2 / H1:M1 N(0,1) . .
--------------------------------------------------

This is the output I get, where I don't get any statistics or p value for cox test and for j test, there is only one model's significance. Can anybody tell me what's going wrong in here, please?

[William Lisowski]

The nntest command is a community contributed command available from SSC.

The output of help nnest is not as helpful as it could be. I believe you want

reg lnprice lnlotsize beddummy1 beddummy2 beddummy3 airco bathrms
nntest lnlotsize bedrooms airco bathrms

Here is a similar example.

. sysuse auto, clear
(1978 automobile data)

. regress price weight length

      Source |       SS           df       MS      Number of obs   =        74
-------------+----------------------------------   F(2, 71)        =     18.91
       Model |   220725280         2   110362640   Prob > F        =    0.0000
    Residual |   414340116        71  5835776.28   R-squared       =    0.3476
-------------+----------------------------------   Adj R-squared   =    0.3292
       Total |   635065396        73  8699525.97   Root MSE        =    2415.7

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
      weight |   4.699065   1.122339     4.19   0.000     2.461184    6.936946
      length |  -97.96031    39.1746    -2.50   0.015    -176.0722   -19.84838
       _cons |   10386.54   4308.159     2.41   0.019     1796.316    18976.76
------------------------------------------------------------------------------

. nnest weight mpg

                   Competing Models                 
  --------------------------------------------------
  M1 : Y = [price]
       X = [weight length]
  M2 : Y = [price]
       Z = [weight mpg]
  --------------------------------------------------
             J test for non-nested models           
  --------------------------------------------------
                    Dist         Stat      P>|Stat|
  H0:M1 / H1:M2    t(70)         1.03       0.305
  H0:M2 / H1:M1    t(70)         2.64       0.010
  --------------------------------------------------
        Cox-Pesaran test for non-nested models      
  --------------------------------------------------
                    Dist         Stat      P>|Stat|
  H0:M1 / H1:M2    N(0,1)       -0.22       0.412
  H0:M2 / H1:M1    N(0,1)      -32.20       0.000
  --------------------------------------------------

.

[Muhammad Ibrahim Shah]

Thank you professor!