- ovtest - also includes the fourth power of yhat as a regressor. thus,
including yhat^4 in your regression should give the same result as ovtest.
I want to know how -ovtest- works.
In my understanding, the RESET procedure is as follows:
1. regress Y on X_1, ... ,X_k, and obtain Yhat.
2. calculate Yhat^2 and Yhat^3.
3. regress Y on X_1, ... ,X_k, and Yhat^2, Yhat^3.
4. test the null that Yhat^2 = Yhat^3 = 0.
But in Stata, -ovtest- does not seem to be identical to this procedure,
as the example below shows.
Does anyone know how -ovtest- is working in Stata?
. sysuse auto
. reg price mpg weight foreign
. predict y
. gen y2=y^2
. gen y3=y^3
. reg price mpg weight foreign y2 y3
. test y2 y3
( 1) y2 = 0
( 2) y3 = 0
Constraint 2 dropped
F( 1, 68) = 3.90
Prob > F = 0.0523
. quietly reg price mpg weight foreign
Ramsey RESET test using powers of the fitted values of price
Ho: model has no omitted variables
F(3, 67) = 15.31
Prob > F = 0.0000
UENOHARA Hideaki, LL.M. <firstname.lastname@example.org>
Doctoral Student, Graduate School of Law and Politics,
The University of Tokyo
* For searches and help try: