Question: Use the PSID1982 data and consider the following two nonnested models (compare with Baltagi 2002, p. 230): M1 : log(wage) = 0 + 1education +
Use the PSID1982 data and consider the following two nonnested models (compare with Baltagi 2002, p. 230): M1 : log(wage) = β0 + β1education + β2experience + β3experience2 +β4weeks + β5married + β6gender +β7ethnicity + β8union + ε M2 : log(wage) = β0 + β1education + β2experience + β3experience2 +β4weeks + β5occupation + β6south +β7smsa + β8industry + ν
(a) Compute the J tests for M1 vs. M2 and M2 vs. M1, respectively.
(b) Both M1 and M2 can be artificially nested within a larger model. Use an F test for M1 versus this augmented model. Repeat for M2 versus the augmented model. What do you conclude?
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
