Prof. MacQueen

 

Purpose 

This assignment provides an opportunity to develop, evaluate, and apply bivariate and multivariate linear regression models.

Resources: 

Microsoft Excel® DAT5/65 Week 5 Data File

Instructions: 

The Excel file for this assignment contains a database with information about the tax assessment value assigned to medical office buildings in a city. The following is a list of the variables in the database:

• FloorArea: square feet of floor space
• Offices: number of offices in the building
• Entrances: number of customer entrances
• Age: age of the building (years)
• AssessedValue: tax assessment value (thousands of dollars)

Use the data to construct a model that predicts the tax assessment value assigned to medical office buildings with specific characteristics.

Construct a scatter plot in Excel with FloorArea as the independent variable and AssessmentValue as the dependent variable. Insert the bivariate linear regression equation and r^2 in your graph. Do you observe a linear relationship between the 2 variables?

Use Excel’s Analysis ToolPak to conduct a regression analysis of FloorArea and AssessmentValue. Is FloorArea a significant predictor of AssessmentValue?

Construct a scatter plot in Excel with Age as the independent variable and AssessmentValue as the dependent variable. Insert the bivariate linear regression equation and r^2 in your graph. Do you observe a linear relationship between the 2 variables?

Use Excel’s Analysis ToolPak to conduct a regression analysis of Age and Assessment Value. Is Age a significant predictor of AssessmentValue?

Construct a multiple regression model.

• Use Excel’s Analysis ToolPak to conduct a regression analysis with AssessmentValue as the dependent variable and FloorArea, Offices, Entrances, and Age as independent variables. What is the overall fit r^2? What is the adjusted r^2?
• Which predictors are considered significant if we work with α=0.05? Which predictors can be eliminated?
• What is the final model if we only use FloorArea and Offices as predictors?
• Suppose our final model is:
• AssessedValue = 115.9 + 0.26 x FloorArea + 78.34 x Offices
• What wouldbe the assessed value of a medical office building with a floor area of 3500 sq. ft., 2 offices, that was built 15 years ago? Is this assessed value consistent with what appears in the database?

2

>Regression Modeling Data

AssessedValue (

000)

790

4
2

796

2

44

4
2
2

4
2

68

3
2

67

00

4
2

2
1
19

2
1

4
2

74

70

2
1
4

3
2
15

2
1
31

3
2
42

1
2
35

2
1
17
678

30

2
2
5

2
1
13

2
2
45

1280
1
1
45

00

3
1

2
2
41

3660
2
2
33

1
2
50

2
2
39

1
1
20

4
3
17

2
2

3
1
5

2
2
15

3
2
27

2
1
8

3
2
17

FloorArea (Sq.Ft.)	Offices	Entrances	Age
							4		8																1
47	20									3	12		1		5
5940	2094
5720	34		19
	3660	38		15
	50		31	1878
2990	949
2610	48	910
5650		42				17
	35	1187
2930	11	13
	1280	671
4880	1	678
1620	710
1820
		45	1585
2570	842
4690	15	39
4	33
	41		27	1268
3530	1251
1094
1110	638
2670	999
1100	653
5810	1914
2560	24	772
2340	890
3690	1282
3580	1264
3610	1162
3960	1447