Objectives: 1. To use a graphing utility to perform a linear regression and use it to make a prediction 2. Write a mathematical model involving direct, inverse, or joint variation • • • • • • • Assignment: P. 110-1: 7a, c, e; 9 a-e; 10 a-e P. 111: 25, 31 P. 112: 39-48 S P. 112: 49-54 S P. 113: 55-62 S, 64, 70 P. 114: 78 HW Supplement Correlation Positive Correlation Negative Correlation No Correlation Correlation Coefficient Line of Best Fit Least Squares Regression Quadratic Regression Inverse Variation Linear Regression Direct Variation Joint Variation Let’s say a set of data consists of two quantities, x and y. In statistics, a correlation exists between x and y if there is a linear relation between x and y. – If y increases as x increases, there is a positive correlation. Let’s say a set of data consists of two quantities, x and y. In statistics, a correlation exists between x and y if there is a linear relation between x and y. – If y decreases as x increases, there is a negative correlation. Let’s say a set of data consists of two quantities, x and y. In statistics, a correlation exists between x and y if there is a linear relation between x and y. – If there is no obvious pattern, there is approx. no correlation. Describe the correlation shown by each scatter plot. A correlation coefficient for a set of data measures the strength of the correlation. -1 ≤ r ≤ 1 • Perfect negative correlation = -1 • No correlation = 0 • Perfect positive correlation = 1 A correlation coefficient for a set of data measures the strength of the correlation. -1 ≤ r ≤ 1 Here are some steps to commit to memory: 1. Press STAT. 2. Under the EDIT menu, choose Edit… 3. Enter your x-values under L1. 4. Enter your y-values under L2. 5. Press 2nd MODE. This QUITS the edit screen. 6. Press 2nd Y=. This is the STAT PLOT menu. 7. Choose Plot1. Highlight On. – Scatter Plot, x-values = L1, y-values = L2. 8. Press ZOOM and choose ZoomStat. The table shows the number y (in thousands) of alternative-fueled vehicles in use in the United States x years after 1997. Graph this data as a scatter plot. Determine if a correlation exists. If a strong correlation exists between x and y, where | r | is near 1, then the data can be reasonably modeled by a trend line. This line of best fit lies as close as possible to all the data points, with as many above as below. More steps to memorize: 1. Press STAT. 2. Under the CALC menu, choose the appropriate regression. 3. Press L1 (2nd 1). 4. Press , (comma). 5. Press L2 (2nd 2). 6. Press , (comma). 7. Press VARS. 8. Under the Y-VARS menu, choose Function…, then Y1. Find the best-fitting line from Exercise 2. Interpret the meaning of the slope and yintercept of your equation. Use your model to predict y when x = 10. The table gives the systolic blood pressure y of patients x years old. Determine if a correlation exists. If it is a strong correlation, find the line of best fit. Then predict the systolic blood pressure of a 16-year-old. • According to your model Y(16)=96, but: Girls (Age 16) Boys (Age 16) 122-132 125-138 • Our prediction was so far off because you can’t reliably extrapolate this far away from the data What your calculator is computing when it finds a line of best fit is the least squares regression. Looking at the image, this occurs where the square of the difference between the actual data and the predicted data is the least. Regression can also be performed on a nonlinear set of data. Quadratic Regression is the process of finding a best-fitting quadratic model for a set of data. This is best done on a calculator! y ax 2 bx c Quadratic Regression Cubic Regression is the process of finding a best-fitting cubic model for a set of data. y ax3 bx 2 cx d Cubic Regression Do your data points keep getting bigger and bigger (or smaller and smaller)? Like exponentially bigger( or smaller)? Try an exponential regression model. y a b x You will be able to model direct variation The simplest kind of line is one that has a slope and intersects the y-axis at the origin. • This type of line shows direct variation. • Equation: y = mx When a line shows direct variation between x and y, we write y = kx, where k is the constant of variation. • y is said to vary directly with x. • k is the same thing as slope! The following are equivalent: 1. y varies directly as (or with) x 2. y is directly proportional to x 3. y = kx, for some nonzero constant k, called the constant of variation or constant of proportionality The variable y varies directly as x. Write and graph a direct variation equation that has the given ordered pair as a solution. Then find each constant of variation. 1. (3, -9) 2. (-7, 4) Since y = kx can be rewritten as k = y/x, a set of ordered pairs shows direct variation if y/x is constant. y = 2x y = 2x + 1 x 1 2 3 4 y 2 4 6 8 x 1 2 3 4 y 3 5 7 9 Great white sharks have triangular teeth. The table below gives the length of a side of a tooth and the body length for each of six great white sharks. Tell whether tooth length and body length shows direct variation. If so, write an equation that relates the quantities. The variable y is directly proportional to the square of x such that y = 18 when x = 6. Write an equation that relates x and y. What is the constant of variation? Find y when x = 12. The variable y varies directly as the square root of x such that y = 18 when x = 9. Write an equation that relates x and y. What is the constant of variation? Find y when x = 16. The following are equivalent: 1. y varies directly as (or with) the nth power of x 2. y is directly proportional to the nth power of x 3. y = kxn, for some nonzero constant k You will be able to model inverse variation A company has found that the demand for its product varies inversely as the price of the product. Write an equation relating demand d and price p. Interpret the meaning of your model. The following are equivalent: 1. y varies inversely as (or with) x 2. y is inversely proportional to x k 3. y = , for some nonzero constant k x Your model could be inversely proportional to the nth power of x • 1 2 2 1 3 2/3 4 1/2 The table compares the area A (in mm2) of a computer chip with the number c of chips that can be obtained from a silicon wafer. Area (mm2), A 58 62 66 70 Number of chips, c 448 424 392 376 1. Write a model that gives c as a function of A. 2. Predict the number of chips per wafer when the area of a chip is 81 mm2. You will be able to model joint variation The following are equivalent: 1. z varies jointly as (or with) x and y 2. z is jointly proportional to x and y 3. z = kxy, for some nonzero constant k Your model could be jointly proportional to the nth power of x and the mth power of y The maximum load that can be safely supported by a horizontal beam is jointly proportional to the width of the beam and the square of its depth, and inversely proportional to the length of the beam. Determine the change in the maximum safe load under the following conditions: 1. The width of the beam doubles 2. The depth of the beam doubles 3. The length of the beam doubles Find a mathematical model to represent this little know statement from classical mechanics: The force of gravity F between two masses m1 and m2 is directly proportional to the product of the two masses and inversely proportional to the square of the distance between them r. Write a sentence using variation terminology for the volume of a cone. 1 2 V r h 3 Objectives: 1. To use a graphing utility to perform a linear regression and use it to make a prediction 2. Write a mathematical model involving direct, inverse, or joint variation • • • • • • • Assignment: P. 110-1: 7a, c, e; 9 a-e; 10 a-e P. 111: 25, 31 P. 112: 39-48 S P. 112: 49-54 S P. 113: 55-62 S, 64, 70 P. 114: 78 HW Supplement