Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

ECA5103 Assignment 1

1. Open the dataset http://fmwww.bc.edu/ec-p/data/wooldridge/crime1 and label all the variables based on the data dictionary http://fmwww.bc.edu/ec-p/data/wooldridge/datasets.list.html.

2. Create a no legal income dummy (no income) that equals 1 if a person does not have any legal income. Generate the summary statistics for narr86, qemp86, inc86, no income, and black and reports the results in Table 1 of the provided template (keep 3 decimal points).

3. Run the following two regressions

and report the regression results in Table 2, and explain the economic meanings of  and .

4. Using the data and the econometrics theory about the potential sign of the bias to explain the observed difference between  and .

5. Generate a new variable log_inc = log(inc86+1) and plot the histogram of inc86 and log_inc

6. Run the following regression

What are the benefits of including no_income in the regression? What are the benefits of using log_inc instead of inc86 as the control variable?

7. Using the formula of Breusch-Pagan test discussed in the class to test the hypothesis that the error term of equation 3 is homoscedastic.

8. If we worried that the income might affect black and non-black people differently, we should use which of the following two models?

Will the of model (4) differ from that of model (5)? What is the predicted difference in narr86 between blacks and non-blacks when the impact is evaluated at the mean income?