The memo(in the case) and regression analysis is full of errors, rule breaking and bad assumptions. Make a bullet point list of at least 15 concerns you have with the analysis or recommendation.

. ELEV 953": V m ., I \"winner 1' No Spac... Heading'l Headingz Title . _ ' , Paragraph _ , ., H Ti ,, Styles In April .2000, Ben M a real estate appraiser in London, Dntario, had received a report from the law firm Penfield Associates. Penfi-eld was involved in litigation regarding the value of 817 Saint Stephen Street in which there had been a dispute about the number of parking spaces included with the property. Mr. What! been hired by Peneld to determine the value of the disputed parking spaces Penfield's opponent, mu Law Partners, hired Hamilton Appraisals who prepared a report using regression analysis. They had assessed the value of the disputed parking spaces at a much higher value than that arrived at by Mrytansipk. Ti? 31? SAINT STEPHEN STREET 817 Saint Stephen was a converted ofce property a residential building that had been converted for commercial use. In December 1999, 817 Saint Stephen St. had been sold to Dr. Michael Mfor use in his optometry practice. There was a parking lot containing five parking spots behind 817 Saint Stephen St, and there were eight additional parking spaces in a lot adjacent to the property. At the time of the sale, Dr. Mbelieved that the lot containing the eight additional spaces was part of 817 Saint Stephen St, however, after the sale he learned that the lot was owned by the City of London. If he wanted to use those spots for his practice, he would have to lease them from the city at a rate of $50 per spot per month. Otherwise, the city would install parking meters in the lot and make it a public facility. THE DISPUTE Dr. Zidar felt that he had overpaid for 817 Saint Stephen Street as it was not made clear to him at the time of the sale that the parking lot was not part of the purchase. He hired the law firm of ggroxall Partners to represent him in his attempt to recover some of the money that he had paid for 817 Saint Stephen St. The property's original owner hired Peneld Associates for the defence, and Peneld hired 'Mr. Wm assess the value of the parking spaces. According to mLapsink, "Everyone agrees that the original owner will have to pay Dr. Zidar. Thetonly question now is how much." REAL ESTATE VALUATIGNS According to Mr. Lansink, theremere three common methods to appraise a property: 1. Determine how much it would cost to recreate the property. For instance, one might estimate how much it would cost to construct a similar building on a siumila rE piece of land. 2. Determine the net present value of the revenue stream generated by the property. 3. Compare the property to similar propertieslthat have recently been sold in the area. [it psaea' Fl _ 54% l' To: Penfield Associates CC: Lansink Appraisals From: Hamilton Appraisals Date: March 25, 2000 Re: 817 Saint Stephen St. Summary We performed a multiple linear regression analysis to determine the value of the eight disputed parking spaces at 817 Saint Stephen St. The regression coefficient for the number of parking spaces indicates that each parking spot would add $11,200 to the value of the property. Thus, the total value of the eight disputed parking spaces is $89,600. Methods We obtained a database of 50 properties sold in London, Ontario, from January 1, 1993, to December 31, 1999. The database included data on the sales price and sales date of each property, as well as many characteristics of the properties, such as square feet, age of building, use of building, etc. The variables included in this database are defined in Table 1. We performed a multiple linear regression analysis to determine the impact of the number of parking spaces on the sales price of a property, while controlling for other factors. We used "sales price" as the dependant variable and treated every other variable as an independent variable. An initial examination of the data revealed that 12 points were outside of the main data cloud. These points were classified as outliers and discarded. To account for potential interactions between the independent variables, an additional variable was defined. The variable park min med was defined as park min med = oum park park min * medical. We performed simple linear regression analyses between sales price and each of the independent variables to determine which independent variables were the most significant predictors of sales price. We then constructed a multiple linear regression model using all those variables found to be significant in the simple linear regression model. The output from the regression model is shown in Table 2. exhibit 1 (continued) The R2 value of the regression indicates that 96 per cent of the variation in sales prices can be explained by the factors included in this multiple linear regression model. A plot of Y versus fitted Y indicates that this is a valid regression model. By examining the p-value for each variable in the model, we conclude that all variables are significant. Since the coefficient of num park is 11157, we conclude that each parking space contributes $11,200 to the value of a property. Table 1: Names and definitions of variables used in the multiple linear regression analysis IAge: Age of the building. Breast: The area of the building in square feet. Brick: Categorical variable; brick = 1 if the building is brick, brick = 0 otherwise. lanes : Number of lanes on the road in front of the building. location : Categorical variable; location = 1 if the location is below average, location = 2 if the location is average, and location = 3 if the location is very good. medical: Categorical variable; medical = 1 if the building is used for medical purposes, medical = 0 otherwise. aww Rack: Number of parking spaces. park min: Categorical variable; park min, = 1 if the property has a minimal amount of parking, park min. = 0 otherwise. sales date Sales date for the property sales price; Sales price of the property. street park: Categorical variable; street park = 1 if there are parking spaces on the street adjacent to the building, street park = 0 otherwise. Use: Categorical variable; use = 1 if the building is used for medical purposes, use = 2 if the building is used for general purposes, and use = 3 if the building is used for retail. Table 2: Summary of output from the multiple linear regression model Variable Coefficient p-value (constant) 675844.4 0.00144 Num park 11157.1 0.00000 age 398.8033 0.00098 area st. 46.172 0.00000 location 21254.33 0.00000 Medical 46407.75 0.00035 Park min 94476.98 0.00000 Park min. med 13843.6 0.00058 Sales date -8785.08 0.00017 Street park -28105.5 0.00011 Use 25207.6 0.00116 R^2=0.957real estate data file Number aadt age area sf brick lanes location pipark 30,800 77 1.969 2 2,100 100 1034 32.900 82 1083 100 6.094 5 20.500 100 2,805 7,700 100 4.380 7 20,500 100 2.953 8 28.800 59 1,083 11, 300 74 3,128 900 10 16.400 45 1.024 11 25.700 100 1,823 12 35.900 59 1,083 13 39.000 18 2,441 14 8,700 64 2,362 15 13,300 65 2,683 16 3,100 90 1.181 17 28,200 43 1.050 18 7,700 59 2,953 19 1,000 57 5.118 W - W N - - NW - W W NW NNNW W - W N W NW - W - NW w - NNW W 20 32,900 100 1,096 21 9,200 73 4.630 22 32,900 46 2,112 23 35,900 1,035 24 27.700 69 1,423 25 17,500 70 2.610 26 33.900 59 2.490 27 8,200 28 1,710 28 100 1.034 29 35,900 54 1.575 30 35,900 63 2.658 31 13,300 69 3.426 32 11,300 100 1,181 33 24.600 58 1,772 34 11,300 91 2.658 35 32,900 100 1.548 36 39,000 4 3.051 37 1,000 53 2.461 38 1,000 58 3.176JX C E F G H Q R Number aadt real estate data file age 1 30,800 area of brick lanes location medica num_pepark_mipark_min # sales_d sales_price street_peuse 77 1,969 9 92456 2 2,100 100 1.034 219 400 3 32,900 82 1.083 153.100 - NNWW 100 6.094 900 5 20,500 2,805 6 7,700 100 4,380 7 20,500 100 2,953 300 8 28.800 59 1.083 9 11,300 74 3,128 10 16,400 45 1,024 11 25,700 100 1.823 12 35,900 59 1,083 13 39,000 18 2,441 14 8,700 64 2,362 15 13,300 65 2,683 16 3,100 90 1,181 17 28,200 43 1,050 18 7,700 59 2,953 CCONG 19 1,000 57 5.118 20 32,900 100 1,096 21 9,200 73 4.630 22 32,900 46 2,112 W - UN - - NW . W W NW - N N NW W . W N - W NW . W - NW_W. 23 35,900 84 1.035 24 27,700 69 1,423 25 17,500 70 2,610 26 33,900 59 2.490 27 8.200 28 1,710 28 100 1,034 29 35,900 54 1.575 30 35,900 63 2.658 31 13,300 69 3,426 32 11,300 100 1,181 33 24.600 58 1,772 34 11,300 91 2,658 35 32,900 100 1,548 NNN-NW 36 39,000 3,051 37 1,000 53 2,461 38 1,000 58 3.176 500