Table 1. Gas Use The gas use data are provided by Timothy Welnetz. The data was collected at his home in Madison, Wisconsin over the dates Sunday January 25, 2004 through Monday February 23, 2004 for a total of 30 time-series observations. Each observation was taken at 8am of the given day and represents the amount of natural gas used (in hundreds of cubic feet) at his home in the previous 24 hours. The purpose of collecting the data was merely to fit various basic time-series models to it and determine quality of fit and prediction for instructional purposes. Gas_Use Gas Source: Timothy Welnetz obs. 30 s: 1 Daily gas use in 100s of cubic feet Table 1 Gas 1 6.1 2 5.0 3 5.0 4 8.4 5 7.1 Table 2. Assessments The home assessments data are provided by Timothy Welnetz. The data was collected from the City of Madison Assessor s Office at http://www.cityofmadison.com/assessor/ in 2005 to compare assessments on the improvements to land (the immovable man-made objects, such as the house and garage, if any) of homes with Lake Mendota water frontage to homes without Lake Mendota water frontage in the Tenney- Lapham neighborhood of Madison, Wisconsin. I collected data on 124 single family residences in this neighborhood focusing only on homes in areas 83 and 28 of the Tenney-Lapham neighborhood and only those homes within one block of Lake Mendota and between N. Ingersoll Street and Baldwin Avenue. The data was not randomly selected, but every home within the defined boundaries above was sampled. Assessments Location Area Assessment obs. 124 s: 3 A categorical variable that denotes whether the property has Lake Mendota water frontage or not Square feet of the living area of the property 2005 assessed value (in thousands of dollars) of the improvements to land Source: City of Madison Assessor s Office, available at http://www.cityofmadison.com/assessor/
Table 2 Location Area Assess 1 No Water 1895 299.2 2 No Water 1934 239.3 3 No Water 1392 207.3 4 No Water 1562 174.4 5 No Water 1440 167.5 Table 3. Ice Cream Sales The ice cream sales data are provided by Timothy Welnetz. This time-series data was made up for instructional purposes to practice correlation and simple regression analysis on data with an outlier. Ice_Cream_Sales Day Temp Sales Source: Timothy Welnetz obs. 20 s: 3 a time index variable High temperature for the day in degrees Fahrenheit Ice cream sales in hundreds of dollars Table 3 Day Temp Sales 1 1 65 20 2 2 68 22 3 3 66 21 4 4 75 23 5 5 81 25 Table 4. House Prices The house prices data was collected by several students for a student project. The data are from houses sold in a community in the state of Oregon (in the early 1990s?). The data was incorporated into a text called Practical Data Analysis: Case Studies in Business Statistics by Peter G. Bryant and Marlene A. Smith first published in 1995. House_Prices obs. 108 s: 2 SQ_FT Square footage of the house PRICE Selling price (in thousands of dollars) of the house Source: Practical Data Analysis: Case Studies in Business Statistics second ed. by Peter G. Bryant and Marlene A. Smith
Table 4 SQ_FT PRICE 1 1238 59.9 2 1707 64 3 1296 66.5 4 1320 66.5 5 1210 69 Table 5. House Prices II The house prices data comes from houses sold in a community in the state of Oregon (in the early 1990s?). The data was collected by several students for a student project. The data was incorporated into a text called Practical Data Analysis: Case Studies in Business Statistics by Peter G. Bryant and Marlene A. Smith first published in 1995. House_II_for _Lab_Practice obs. 108 s: 11 SQ_FT Square footage of the house BEDS bedrooms BATHS bathrooms HEAT A binary variable 0 is gas forced air heating and 1 is electric heat STYLE Architectural style of the home: 0 is two-story, 1 is tri-level, 2 is ranch GARAGE cars that can fit into the garage BASEMENT A binary variable 0 is no basement; 1 is there is a basement AGE Age of the home in years FIRE A binary variable 0 is no fireplace, 1 is there is at least 1 fireplace PRICE Selling price (in thousands of dollars) of the house SCHOOL A binary variable 0 is Eastville school district and 1 is Apple Valley school district Source: Practical Data Analysis: Case Studies in Business Statistics second ed. by Peter G. Bryant and Marlene A. Smith Table 5 SQ_FT BEDS BATHS HEAT STYLE GARAGE BASEMENT AGE FIRE PRICE SCHOOL 1 1238 3 2 0 0 1 1 12 59.9 1 2 1707 3 2 1 0 2 0 13 64 0 3 1296 4 2 0 0 2 1 17 66.5 0 4 1320 3 2 0 0 2 1 11 66.5 0 5 1210 3 2 0 0 1 0 6 69 0
Table 6. Premium The premium data was collected by several students for a student project. The data represent characteristics of 86 drivers and their relationship to how much they pay for their car insurance. The data set and corresponding correlation and regression analysis was incorporated into a module called Selection as part of a larger compilation called the Data Analysis Handbook written by Marlene A. Smith. Premium Age_Drvr Age_Car Tickets Discount Female Premium obs. 86 Source: Data Analysis Handbook by Marlene A. Smith s: 6 Age of the policyholder in years Age of the insured vehicle in years traffic violations received by the policyholder within the last three years discounts applied to the policy Sex of the insured - binary variable coded 1 for females and 0 for males The amount paid for six months of insurance coverage Table 6 Age_Drvr Age_Car Tickets Discount Female Premium 1 75 2 0 2 0 391.9 2 23 6 0 2 1 551.1 3 74 3 2 2 0 654.2 4 24 5 0 2 1 517 5 21 1 0 3 0 1067.6 Table 7. Retail Sales The Retail Sales data was retrieved by Timothy Welnetz from the US government s census website at http://www.census.gov/retail/mrts/historic_releases.html. The data collected is monthly U.S. retail sales including food sales (seasonally unadjusted) in billions of dollars beginning January 2001 and ending June 2006 for a total of 66 time series values. The purpose of collecting the data was merely to fit various basic timeseries models to it (including several that incorporate seasonal components) and determine quality of fit and prediction for instructional purposes. Retail_Sales Mo./Yr. obs. 66 s: 2 Month and Year the value corresponds to
Sales Source: http://www.census.gov/retail/mrts/historic_releases.html US Retail Sales in billions of dollars Table 7 Mo./Yr. Sales 1 Jan-01 251.059 2 Feb-01 248.078 3 Mar-01 280.749 4 Apr-01 275.160 5 May-01 296.459 Table 8. Pulse Rates The pulse rate data was collected by Timothy Welnetz for instructional purposes to demonstrate regression analysis concepts in the context of a designed experiment. The data set consists of thirty pulse rate readings taken from a heart rate monitor at the end of each minute walking on a treadmill. In addition, two factors were varied randomly for each minute: treadmill speed, ranging from 3.6 to 4.8 miles per hour and treadmill incline, ranging from no incline to an incline at a setting of 8. The maximum incline setting was 15. Pulse_Rates obs. 30 s: 4 Period The time period (minute) on the treadmill in which a pulse reading was taken Incline Incline of the treadmill (0, 4, or 8) Speed Walking speed on the treadmill (3.6, 3.9, 4.2, 4.5, or 4.8) Pulse Pulse rate (beats per minute) Source: Timothy Welnetz Table 8 Period Incline Speed Pulse 1 1 0 4.2 115 2 2 0 4.8 125 3 3 8 4.5 138 4 4 4 3.6 127 5 5 0 4.5 124 Table 9. Start Renting 2003 The Start Renting 2003 data was collected by Timothy Welnetz from a Madison, Wisconsin apartment rental resource magazine called Start Renting from a January, 2003 issue. The data collected consists of three apartment characteristics for apartments located in Madison, Wisconsin. The primary reason for collecting this data was to see and model the relationship between two metric variables while incorporating a third binary variable.
Start_Renting_2003 Location Sq Ft Rent obs. 200 Source: Start Renting Magazine, January 2003 issue s: 3 Area of Madison where the apartment is located: South or West Square footage of the apartment Monthly rent for the apartment Table 9 Location Sq Ft Rent 1 South 353 440 2 South 430 490 3 South 450 525 4 South 500 480 5 South 550 515 Table 10. Start Renting 2010 The Start Renting 2010 data was collected by Timothy Welnetz from a Madison, Wisconsin apartment rental resource magazine called Start Renting from an August, 2010 issue. The data collected consists of fifteen apartment characteristics for apartments located in Madison, Wisconsin. The data can be used to model characteristics of apartments that relate to apartment rent, for example. Start_Renting_2010 obs. 256 s: 16 Apartment Apartment identifier Location Area of Madison where the apartment is located: West, Near West, Central, Near Eastm North, Northeast, Southeast, Southside Type Type of Apartment: Room, Studio, Efficiency, or 1 to 6 Bedroom Rent Monthly rent (in dollars) Size 97 Size in square footage Baths bathrooms Heat Whether heat is included: 1=yes, Elec Whether electricity is included: 1=yes, Air Whether air conditioning is included: 1=yes, Water Whether water is included: 1=yes,
Appl Laund Garage Bus Cats Dogs Source: Start Renting Magazine, Aug. 19 Sept. 1, 2010 issue Whether appliances are included: 1=yes, Whether laundry facilities are onsite: 1=yes, Whether a garage is available for parking: 1=yes, Whether the apartment is located on a busline or not: 1=yes, Whether cats are allowed: 1=yes, Whether dogs are allowed: 1=yes, Table 10 Apartment Location Type Rent Size Baths Heat Elec Air Water Appl Laund Garage Bus 1 1 West Room 700 1 1 1 1 1 1 1 0 1 2 2 West Studio 476 400 1 0 0 1 1 1 1 0 1 3 3 West Studio 505 340 1 0 0 1 1 1 1 0 1 4 4 West Studio 524 485 1 1 0 1 1 1 1 0 1 5 5 West Studio 535 1 0 0 1 1 1 1 0 1 Table 10 Example of the first five observations continued: Cats Dogs 1 0 0 2 1 0 3 1 1 4 1 0 5 0 0