The RAMSEY RESET TEST is carried out when the tester feels that the functional form may not be right. For instance, a general linear regression will take the form as shown below:
However, the tester may feel that the functional form should take the form of :
or something else which is non linear. As such, the RAMSEY RESET test is one test that can be used to detect omitted nonlinearity. The test can be carried out as shown below:
Note that the null in this test means that the non linear terms cannot help to explain the dependant variable. If the null is rejected, it means that the non linear terms are in fact important and thus it should be included in the regression model.
No comments:
Post a Comment