Suppose we are interested in the effect of school rankings on law school graduates earnings. Further...

 

Transcribed Image Text:

Suppose we are interested in the effect of school rankings on law school graduates earnings. Further suppose that we are unsure of the true model so we estimate two models using random sample: 0.0041 rank +0.294 GPA (0.0003) (0.069) Isalary 9.90 (0.24) n = 142, R = 0.8238 Isalary = 9.86 -0.0038 rank +0.295 GPA+0.00017 age (0.26) (0.0004) (0.083) (0.00036) n = 99, R = 0.8036 (1) (2) where Isalary is the log salary of the graduate, rank is the ranking of the school they graduated from, GPA was the cumulative GPA, and age is their age. (a) (2 points). Interpret the estimate of the coefficient of rank in model (1). (b) (2 points). Interpret the R of model (1). Would you say most of the variation in log(salary) is explain by the independent variables? (c) (4 points). Does including age greatly affect the other parameters estimates? Would you say that the zero conditional mean assumption on the error will hold in the first model? Explain. (d) (4 points). Does including age greatly affect the standard errors? Why do you think? What else could have caused the change? (e) (2 points). Which model is more likely to have OLS estimators that satisfy the Gauss-Markov assumptions? Suppose we are interested in the effect of school rankings on law school graduates earnings. Further suppose that we are unsure of the true model so we estimate two models using random sample: 0.0041 rank +0.294 GPA (0.0003) (0.069) Isalary 9.90 (0.24) n = 142, R = 0.8238 Isalary = 9.86 -0.0038 rank +0.295 GPA+0.00017 age (0.26) (0.0004) (0.083) (0.00036) n = 99, R = 0.8036 (1) (2) where Isalary is the log salary of the graduate, rank is the ranking of the school they graduated from, GPA was the cumulative GPA, and age is their age. (a) (2 points). Interpret the estimate of the coefficient of rank in model (1). (b) (2 points). Interpret the R of model (1). Would you say most of the variation in log(salary) is explain by the independent variables? (c) (4 points). Does including age greatly affect the other parameters estimates? Would you say that the zero conditional mean assumption on the error will hold in the first model? Explain. (d) (4 points). Does including age greatly affect the standard errors? Why do you think? What else could have caused the change? (e) (2 points). Which model is more likely to have OLS estimators that satisfy the Gauss-Markov assumptions?

Answer rating: 100% (QA)

a Interpretation of the coefficient of rank in model 1 The estimated coefficient of rank in model 1 ... View the full answer