Q1. The tab labeled "Problem 1" includes monthly price data on the S&P 500 (SPY) as well as on KB Home (KBH, a real estate development company) and on Camden Property Trust (CPT, an equity REIT specializes in apartment rentals). Using regression, calculate the Beta of each of these two real estate related companies for the full time period provided to you.

a. Display your calculated Betas in the cells highlighted in yellow. You may include your full regression results in two other separate tabs.

b. Which firm has a higher Beta? In one to two sentences explain why you did or didn't expect that firm to be associated with a higher Beta.

Q2. The tab labeled "Data" includes data on recent single family sales in Palm Beach County, FL. Using the data provided, follow the instructions and answer the questions listed below:

Filter and copy into the tab labeled "Filtered Data" the observations that meet the following criteria:

• Only properties with SqFtLA (SQFT living area) of 1,500 to 4,000.
• Only properties located in Boca (Boca Raton) or Delray
• Only properties with reasonable Sale Price, number of bedrooms, full bathrooms and half bathrooms (#Beds, #FB and #HB). Please filter out all properties for which the #HB cell is blank.

In the same tab (Filtered Data) create the following variables.

• Baths - A variable that combines #FB and #HB into the total number of bathrooms. For example, a property with 2 full baths and 1 half baths would be converted into 2.5 baths.
• Age - A variable that uses the closing date (CD) and the year built (Yr Blt) in order to determine how old the property was (in years) when it was sold.
• PoolDum - A pool dummy variable that is set to 1 if the property has a pool (Pool? is Y) and 0 otherwise.
• BocaDum - A location dummy variable that is set to 1 if the property is located in Boca (CITY is BOCA) and 0 otherwise.
• LnPrice - A variable that includes the natural log of the Sale Price.
• Lnsqft - A variable that includes the natural log of SqFt LA.

Run the following two regressions and answer the questions below within the yellow cells included in the "Problem 2" tab:

(1) SalePrice = α + β₁SqFtLA + β₂#Beds + β₃Baths + β₄Age + β₅Pool + β₆Boca

(2) LnPrice = α + β₁Lnsqft + β₂#Beds + β₃Baths + β₄Age + β₅Pool + β₆Boca

a. What is the R^2 of regression (1) and regression (2)?

b. Why do you think that the R^2 on these regressions is lower than the R^2 of the examples we used in class? Is there a particular variable or factor that you think would increase the R^2 in a material way? Briefly explain.

c. On average, by how much (in dollars) one square foot would increase or decrease the property selling price?

d. On average, by how much (in percentage) a pool would increase or decrease the property selling price?

e. On average, by how much (in percentage) a property in Boca would sell in comparison to a property in Delray?

f. Is the coefficient of Age positive or negative? Briefly explain the reason behind this positive or negative coefficient.

Q3. Please refer to the paper titled "The Effect of Listing Price Strategy on Transaction Selling Prices" (file name "Beracha Seiler_JREFE 2013.pdf" under Meeting 2 on BlackBoard) and answer the following questions:

a. According to Table 2, what is the most common housing pricing strategy employed in the market? What is the second most common pricing strategy?

b. If every property would be marketed exactly at the price that equals its value, what would be the percentage associated with the "Thousands digit" 7 in Table 2? Briefly explain why.

c. According to Table 3, which housing price strategy is associated with the smaller price discount? Is this expected? Briefly explain why.

d. Briefly explain the main point that we can learn from the regression results presented in column (4) of Table 6.

e. Do the results presented in Table 6 conflict with the results presented in Table 3? Briefly explain.

Q4. Assume that you have a dataset that includes the following variables on each commercial real estate transaction that took place during the 2010-2016 time period in Omaha, NE and in Minneapolis, MN (yes, they do have commercial real estate in these cities.....)

- YR - The year the transaction took place.

- Price - The price associated with the transaction.

- SQFT - The size of the property transacted in sqft.

- Income - The total annual rent collected during the year before the transaction.

- Class - A value between 1 and 3 denoting the property class, where 1, 2 and 3 represent classes A, B, and C, respectively.

- Location - A value of 0 or 1 denoting the property location, where 0 and 1 represent Omaha and Minneapolis, respectively.

Assume that the following regression equation was applied on the data and answers the questions below:

Price = α + β₁SQFT + β₂YR + β₃Income + β₄Class + β₅Lacation

a. The coefficient β₃ is positive and statistically significant. Is that expected? In one sentence explain why.

b. The coefficient β₂ is positive and statistically significant. Is that expected? In one sentence explain why.

c. Would you expect the β₄ coefficient to be positive or negative? Would you expect it to be statistically significant? In one sentence explain why.

d. Would you expect the β₅ coefficient to be positive or negative? Would you expect it to be statistically significant? In one sentence explain why.

e. The coefficient β₁ is negative and statistically insignificant. Does it make sense? In one sentence explain why.

Q5. Referring to problem 4, if you could get any information you want about each property:

a. In a similar way to how the variables are defined in problem 4, define two other variables that you would have liked to have included in your regression in order to increase the estimation power of the regression.

b. For each of the variables you defined in part a, briefly explain the sign you would expect (negative, positive or ambiguous) from the coefficient of each of these variables.

Attachment:- Assignment Data.rar