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

Problem Set

You must turn in a PDF with your answers and a script (do file).

To get full marks, your code must run without errors.

The problem set is designed assuming you will run your code in Stata.

For this problem set, you will use the dataset called data_ps1.dt

1. Document the relationship between y1 and x1

a. Plot the scatter plot of y1 and x1

b. Run the linear regression

c. What can you say about the relationship between y1 and x1

2. Document the relationship between y1 and x1 controlling for x2 and x3.

a. Plot the scatter plot of y1 and x1 controlling for x2 an x2

b. Run the linear regression

c. Is it possible to recover the coefficient on x1 from point 2b in a different way? How?

3. Now, study the relationship between y2 and x1

a. Plot the scatter plot of y2 and x1

b. Run the linear regression

c. Is the model misspecified? If so, offer diagnostic and suggest corrections

4. Now, we discover that our dataset contains a cluster_var. Let’s study the relationship between y0, x0, and x1, paying close attention to the computation of std errors.

a. Compute homoscedastic std errors using the canned routine

b. Compute the same errors manually

c. Compute the Eicker-Huber-White std errors using the canned routine

d. Compute the same errors manually

e. Compute the Liang-Zeger cluster standard errors using the canned routine

f. Compute the same errors manually