ECON 4750代写、R编程语言代写、代做website留学生、代写R设计

日期：2018-12-06 09:39

William B. Vogt ECON 4750 (13-680) Final

Final

Assigned: December 4, 2018

Due: 11:30 pm, December 12, 2018

Instructions:

Please turn in a pdf document with your answers in it. Please turn in

a separate document with your R code in it. Turn in the document

on the class elc website.

You may NOT cooperate or seek help from anyone (other than the

course instructor).

Please be sure to include your R program/script as an appendix to

your test (i.e. as a separate file). I should be able to copy and paste

the script from your document into R, and it should reproduce the

results you generated to answer the test questions.

Please remember that presenting estimates of uncertainty is part of

answering questions with data. Even if I do not prompt you to do so

in the body of the question itself.

Make sure you use the tools developed in class (and not your random

gut feelings)to answer the questions.

Terry College 1 of 3 UGA

William B. Vogt ECON 4750 (13-680) Final

Problems: Housing Prices In Windsor, Canada

There is a dataset of housing prices and other information for Windsor,

Canada in the R library AER. You will have to install the package AER.

Then, to get access to the data, issue the command data("HousePrices").

You can find rudimentary documentation on the data by issuing the command

help("HousePrices").

The dataset contains prices and other characteristics for houses sold in

Windsor, Canada in 1987. We will focus on understanding how housing

characteristics affect these prices. Specifically, we want to know how the

size of the lot affects the price of the house.

1. (10 points) Graph price against lot size. What does the relationship

look like?

2. (15 points) Write down a linear regression model to estimate how price

varies with the lot size. Estimate the model. Interpret your results

focusing on how much an extra square foot of lot size increases price.

3. (20 points) A critic complains that big houses are often built on big

lots and that your answer in the previous problem does not account for

this. Evaluate his criticism (what assumption is he saying is violated,

what consequences, does it seem he is right). If necessary, estimate a

new regression to fix any problem identified by this critic and answer

the previous question again with it.

4. (20 points) A critic complains that he thinks the relationship between

house price and the right-hand-side variables should be nonlinear.

Evaluate and if necessary correct for this criticism. Interpret any new

regression you run.

5. (15 points) You now decide that you are interested in what the effect

of a good neighborhood is on housing prices. To this end, you estimate

a log-log regression of price on lotsize, number of bedrooms, number of

bathrooms, and a dummy for the preferred neighborhood. How much

higher are prices in the preferred neighborhood, c.p.?

6. (10 points) Predict the price of a house with 3 bedrooms and 2 bathrooms

on a 5000 square foot lot not in the preferred neighborhood.

7. (10 points) A critic complains that the error term is heteroskedastic.

Answer question 5 again, correcting for this problem (you will have to

Terry College 2 of 3 UGA

William B. Vogt ECON 4750 (13-680) Final

look at the R example 8 on the website which shows how to do this

using vcovHC).

Terry College 3 of 3 UGA