During the 2020 Covid-19 pandemic, home prices increased significantly due to the increase in demand and the limited supply. According to a report by Zillow, in May 2021, the average home price increased by 13.4% compared with May 2020. This is a record increase in home prices since Zillow started collecting this data in 1996. So, what determines home value? Is it location? Is it size? Is it the number of bedrooms? Is it yard size? Is it the age of the house? There are many factors that affect the value of a home. Use the Home Price 1 Final Project Data and Home Price 2 Final Project Data files on Campus Web to examine the factors affecting home values and write a 6 to 10-page report with your findings.

1. Descriptive Analytics: The first step in analyzing any data set is to convert it into a more meaningful way, so that you can get the information you need easily and quickly. The Home Price 1 Final Project Data file posted on Campus Web contains data on house prices along with other variables related to houses. Eastern Realtor, Inc owns a rental property in a few college towns across the U.S. and wants to expand its holding through buying another rental property in either Athens, Georgia, or Chapel Hill, North Carolina. They would like to get information about house prices in these two cities. They are interested in properties that have 2 or more bedrooms and are listed for less than $1 million. Use the Home Price 1 Final Project Data file to provide them with this information. Make sure you include the following:

a. Data summary statistics for home prices in the two cities and comment on it.

b. Present house prices in each city using different data visualization tools. Make sure you include both tables and graphs and comment on each one.

c. The boxplot of house prices in each city and comment on it. According to your boxplot, are there any outliers in house prices in any of these two cities?

d. Compare house prices in both cities, what conclusion(s) do you get?

2. Predictive Analytics: Regression Analysis: Eastern Realtor, Inc. is now considering the college town of Ames, Iowahome to Iowa State University. Eastern is looking at newer and older houses in Ames, IA. Newer houses are the ones built in year 2000 or after, while older houses are the ones build before 2000. Build a predictive model to forecast house prices in Ames, IA and make sure you include a dummy variable to take care of the new/old house. Make sure you include the following:

a. Descriptive statistics for all houses, new, and old houses in Ames, IA and comment on it.

b. How many old houses and how many new houses are in Ames, IA?

c. Is there a difference in size between the newer and old houses in Ames, IA? According to a 2017 article in Building magazine, newer houses have become 24% larger than older houses over the past 15 years, does your finding for Ames, IA agree with this article?

d. Build a linear model with the predictors given in the data file and the dummy variable to account for whether the house is new or old. Are all your variable significant?

e. In your model, do you think there is a multicollinearity between any variables? If so, you may want to drop one variable and estimate another model without it. Are all your variables significant now?

f. Compare the two models and choose the best one.

g. Explain the estimated coefficients in your best model.

h. According to your estimated model, compare the predicted price of a 1600-square-foot house with 2 bathrooms and 15,000-square-foot lot for a newer house and for an older house.

3. Predictive Analytics: Time-Series Forecasting: Home Price 2 Final Project Data file on Campus Web contains data on median home prices in the West Census region quarterly from 2010 to 2018. Easter Realtor, Inc. is interested in houses in this region. They want to use this data to forecast home price in the fourth quarter of 2018. For cross-validation, use the training set from the first quarter of 2010 to the fourth quarter in 2016, and use a validation set from the first quarter of 2017 to the third quarter of 2018. Use this data to do the following:

a. Use the training set to build a linear and quadratic trend models with seasonal dummy variables to capture the seasonality.

b. Calculate the RMSE (or the MSE) for the validation set of each model and choose the best model.

c. Re-estimate the best model using the entire data set and use it to forecast home price in the fourth quarter of 2018.