AFM Unit 9 Regression Day 1 notes

A mathematical model is an equation that best describes a particular set of paired data.These mathematical models are referred to as __________________________ models and are used to _____________one variable based upon another variable.

Linear Regression: STAT CALC #4 _____________________

Correlation Coefficient: the quantity _____, measures the strength and direction of a linear relationship between 2 variables.

__________≤r ≤¿ If x and y have a strong positive linear correlation, r is close to _______

If a perfect positive fit , then r = ________. (the slope of the line will be ____)

Positive values: as values for x increases, values for y ____________.

Negative correlation: r will be close to _______. An r value of -1 indicates a _____________ negative fit. (slope = ___)

Negative values: as values for x increase, values for y __________________.

No correlation: If there is no linear correlation or a very weak linear correlation, r is close to ______________.

Interpolation: Predicting values with an equation of best fit _________________ the range of data

Extrapolation: Predicting values with an equation of best fit ______________ the range of data.

Example 1. Is there a relationship between Math SAT scores and the number of hours spent studying for the test?A study was conducted involving 20 students as they prepared for and took the Math section of the SAT Exam._______________________________________a.) Determine a linear regression model

equation to represent the data._____________b.) Graph in the calculator and decide whether the equation is a “good fit”.

_____________c.) If a student studied for 15 hours, what is the expected Math SAT score?

______________d) Is part c an example of interpolation or extrapolation?

______________e) If a student obtained a Math SAT score of 720, how many hours did the student most likely spend studying?

______________f) If a student spent 100 hours studying, what would be the expected Math SAT score?

______________g) Is part “f” an example of interpolation or extrapolation?

Example 2.

_________________________________a) Determine a linear regression model equation to represent this data.

_________________________________b) Graph and determine whether the new equation is a “good fit”.

________________________________c) If the ground temperature reached 95⁰F, then at what rate would you expect the crickets to be chirping?

_________________________________d) Is part c an example of interpolation or extrapolation?

________________________________e) With a listening device, you discovered that on a particular morning the crickets were chirping at a rate of 18 chirps per second. What was the approximate ground temperature that morning?

_________________________________f) Is part “e” an example of interpolation or extrapolation?

_________________________________g) If the ground temperature should drop to freezing (32⁰F), what happens to the cricket’s chirping rate?

AFM Unit 9 Linear regression Day 1 NAME _________________________Round all answers to nearest ten thousandths.

1. The table shows statistics for nine NHL career scoring leaders.

Games Played, x 1487 1767 1269 1348 1282 1432 1756 1425 1514Goals Scored, y 894 801 741 731 717 708 694 692 692

_____________________________________________________a. Find an equation of the least-squares line.

____________________b. Find the correlation coefficient.

____________________c. Mike Modano played 1499 games. Predict the number of goals he might have scored.

2. Brian wanted to determine the relationship that might exist between speed and miles per gallon of an automobile. Let x be the average speed of a car on the highway measured in miles per hour and let Y represent the miles per gallon of the automobile. The following data are collected.

X 50 55 55 60 60 62 65 65Y 28 26 25 22 20 20 17 15

______________________________________a. Use a graphing calculator to find the line of best fit.

______________________________________b. Interpret what the slope means.

______________________________________c. Predict the miles per gallon of a car traveling 61 miles per hour.

______________________________________d. Predict the speed of a car that gets 24 miles per gallon.

3. A doctor wished to determine whether a relationship exists between the height of a female and weight. She obtained the heights and weights of 10 females aged 18-24. Let height be the independent variable, X, measured in inches, and weight be the dependent variable, Y, measured in pounds.

X 60 61 62 62 64 65 65 67 68 68Y 105 110 115 120 120 125 130 135 135 145

__________________________a. Use a graphing calculator to find the line of best fit.

__________________________b. Interpret what the slope means.

__________________________c. Predict the weight of a female aged 18 to 24 whose height is 66 inches.

__________________________d. Predict the height of a female whose weight is 122 pounds.

For each pair of variables, tell whether you think the correlation is positive, negative, or approximately zero. Briefly give your reasons.4. Average daily temperature in Waterville during January and average daily heating cost in Waterville during January.5. A person’s income and the value of his or her house.6. The amount of annual taxes a person pays and his or her height.7. The height of a person and weight of a person8. The height of a person and the average height of his or her parents9. The value of an automobile and its age

10. The incident of flu and the outside temperature11. The height of a person and the number of years of formal education that person has completed.12. Six Scatter plots are shown. Match each scatter plot to one of the following estimated values of r:

0.8, 0.6, 0.2, 0, -0.2, -0.6, -0.8.

13. In a ten-week course, the final grades A, B, C, D, and F were given the scores y = 4, 3, 2, 1, and 0, respectively. The linear regression equation for predicting scores from x, the number of absences, was

. What letter grade can you predict for a student with

a. 1 absence? b. 3 absences? c. 8 absences?

14. When the sales volume in hundreds of units is plotted against x, the money spend on advertising in thousands of

dollars, researchers obtained a linear regression equation What average sales volume (in hundreds of units) can you predict for the following amounts (in thousands of dollars) spent on advertising?

a. 10 b. 5 c. 7

15. Use the table to answer the following questions.X = amount of nitrogen (pound per acre) 111 145 16

9199 230 251 287 304 335 387

Y = amount of alfalfa (tons per acre) 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5

______________________________a. What is the linear regression equation?_________________b. What is the correlation coefficient?________________c. What amount of alfalfa is present when then amount of nitrogen is 150 pounds per acre?________________d. What is the amount of nitrogen when the amount of alfalfa is 7 tons per acre?

16. Use the table to answer the following questions.X = standardized test math score 400 450 500 500 550 600 650 700 750 800Y = college first-year math avg. 50 60 65 70 78 80 84 88 92 95

___________________________________________________a. What is the linear regression equation?

___________________________b. What is the correlation coefficient?

___________________________c. What would be the college first-year math average for a student that scored a 575 on the standardized math test?

___________________________d. What would the score on the standardized math test be for a student that had an 88 college first-year math score?

b. linear or

exponential?______________________________c. write the equation.___________________d. price of a movie ticket in 2000? ___________________e. price of a movie ticket in 2010?

a. Find the correlation coefficient for each type of equation. (linear, exponential) round to 4 places____________________, _____________________________________________b. Which equation best fits the data?

5.

AFM Unit 9 Day 2 HW Exponential Regression Worksheet

1. Fit an exponential curve for each set of data. Find the correlation coefficient.a.

x 1 3 4 5y 3.00 6.75 10.13 15.19

b.

x 2 2.5 4 5.5y 3.63 3.81 4.39 5.07

2. Each year after he bought his new car, Mr. Brown kept track of the market value of the car.

x = year after purchase 1 2 3 4 5 6y = market value (in dollars) 12,000 9600 7700 6200 4900 3900

a. Fit an exponential curve to the data.b. What is the correlation coefficient?c. Based on the correlation coefficient, is the model a good fit to the datad. Predict the value of the car when it is 10 years old.e. Estimate the amount that Mr. Brown paid for the car.

3. A cup of hot tea just poured at 158℉ slowly cools over time t (in minutes) and its temperature T is recorded.

t 0 10 30 50 70 90 110 120 125 130T 158 132.8 105.8 92.3 84.2 79.2 76.1 75 74.7 74.5

a. Fit an exponential curve to the data.b. What is the correlation coefficient?c. Based on the correlation coefficient, is the model a good fit to the data?d. What will the temperature of the tea be after 60 minutes?e. What will the temperature of the tea be after 150 minutes?f. Estimate the room temperature.

4. Each student in a typing class is tested at various times in the course and the average number of errors for the class is recorded.

T = time of testing (days) 2 10 14 21 30 45 63 70 91Y = average number of errors 45.2 36.

130.2 23.1 18.7 11.0 5.6 4.3 2.4

a. Fit an exponential curve to the data.b. What is the correlation coefficient?c. Based on the correlation coefficient, is the model a good fit to the data?

d. What is the average number of errors for a student after 50 days?e. How many days would it take a student to practice to only have 10 errors?

AFM Unit 9 Day 3 HW Power Regression and Choosing Best Model WorksheetRound all answers 4 places after the decimal.1. Fit a power curve for each set of data. Find the correlation coefficient and the equation.a. b.

2. In 1610, Galileo discovered how the time, T, required for each of Jupiter’s satellites to revolve about Jupiter is related to the average distance, a, from Jupiter.

____________________a. Find the power regression equation._____________________b. What is the correlation coefficient?_____________________c. Is the power model a good fit for the data?

3. The winning times (in seconds) for various men’s races (in meters) in the 1988 Olympics are given.x = distance (m) 100 200 400 800 1500 5000 10,000y = time (seconds) 9.92 19.75 43.87 103.45 215.96 791.70 1641.46

_________________________________________________a. Find the power regression equation._________________________________________________b. What is the correlation coefficient?_________________________________________________c. Is the power model a good fit for the data?_________________________________________________d. What would the predicted winning time be for a 1000 meter race?___________________________________________e. For the winning time to be 500 seconds, how long would the race have to be?

4. A scientist caught, measured, weighed, and then released eight Maine landlocked salmon. x = length (in.) 5.5 10.6 15 17 19.6 22 25 28y = weight (lb) 0.1 0.4 1.0 1.6 2.5 3.5 5.4 7.4

__________________________________a. What is the correlation coefficient for a linear model for this data?__________________________________b. What is the correlation coefficient for a exponential model for this data?__________________________________c. What is the correlation coefficient for a power model for this data?__________________________________d. Which model best fits the data?_____________________________________________________________e. What is the best model equation?____________________________f. Predict the weight of a salmon that is 12 inches long.5. The table shows the average salaries (in thousands of dollars) of major league baseball players for selected years from 1967-1989.

_________________________________a. What is the correlation coefficient for a linear model for this data?_________________________________b. What is the correlation coefficient for a exponential model for this data?_________________________________c. What is the correlation coefficient for a power model for this data?_________________________________d. Which model best fits the data?_____________________________________________________________________________e. What is the best model equation?_________________________________f. Estimate the average salary in 1970.

6. Various depths (in meters) below the water surface and corresponding water pressures (in atmospheres) at those depths are given in the table. (one atmosphere is about 14.7 lb/in.2)

x 2.0 3.0 5.0 6.0y 4.0 13.5 62.5 180.0

x 0.9 1.5 2.4 3.0y 4.94 1.78 0.70 0.44

Satellite a (kilometers) T (hours)Io 422,000 42.5Europa 671,000 85.2Ganymede 1,072,000 171.7Callisto 1,883,000 400.5

year 1967 1976 1979 1982 1984 1986 1989salary (in thousands) 19.0 51.5 114 242 329 413 497

d = depth (m) 10 20 30 40 50 60P = pressure (atmospheres) 15.0 31.0 44.5 60.1 74.8 88.1

____________________________a. What is the correlation coefficient for a linear model for this data?____________________________b. What is the correlation coefficient for a exponential model for this data?____________________________c. What is the correlation coefficient for a power model for this data?____________________________d. Which model best fits the data?_______________________________________________________________________________e. What is the best model equation?____________________________f. Predict the water pressure at a depth of 13 m.