Question

b- We have a second estimation in which we include a third explanatory variable: rank of the law faculty (being ???? = 1 the best one) such that:

log(????) ̂ i = 6.34 + 0.095 log(????)i + 0.38 log(????)i − 0.0033????? ? = 200 ? 2 = 0.294

Why ????? variable is not included in the model in logs? Which is the model with a better goodness-of-fit? Do these university specific variables explain the behaviour of the wage variable? Why

Answer 1

With available information in this specific part (b)

Ranki is not inculcated as log because its a dummy variable that takes only value either 1 or 0. And if the value is the value is 1 then ln(1) is zero. So its is same as not including the variable .

For better goodness of fit we can compare R-2 values of two regression models more the value more the goodness of fit of the model.

To check whether university specific variable is able to expain the variation in wage or not we can check significance of the Ranki coefficient. But do Not availability of standard errors of coefficient.

If it is significant then we can say that yes this variable can explain the variation in wage.

And this may be because more is rank of university more we will expect the student's education quality and then more is the wage .

Sorry for improper answers because required information is not provided in part (b)