2. 1 Simple Linear Regression Review 1. review of scatterplots and correlation 2. review of least squares procedure 3. inference for least squares lines

3. 2

4. Basic Terminology Univariate data : 1 variable is measured on each sample unit or population unit e.g. height of each student in a sample Bivariate data : 2 variables are measured on each sample unit or population unit e.g. height and GPA of each student in a sample; ( caution: data from 2 separate samples is not bivariate data)

5. Basic Terminology (cont.) Multivariate data: several variables are measured on each unit in a sample or population. (later in course)  For each student in a sample of NCSU students, measure height, GPA, and distance between NCSU and hometown. Focus on bivariate data for now.

6. 5 Introduction We will examine the relationship between quantitative variables x and y via a mathematical equation. The motivation for using the technique: –Forecast the value of a dependent variable (y) from the value of independent variables (x 1, x 2,…x k.). –Analyze the specific relationships between the independent variables and the dependent variable.

7. Scatterplot: Fuel Consumption vs Car Weight. x=car weight, y=fuel cons. (x i, y i ): (3.4, 5.5) (3.8, 5.9) (4.1, 6.5) (2.2, 3.3) (2.6, 3.6) (2.9, 4.6) (2, 2.9) (2.7, 3.6) (1.9, 3.1) (3.4, 4.9)

8. The correlation coefficient is a measure of the direction and strength of the linear relationship between 2 quantitative variables. It is calculated using the mean and the standard deviation of both the x and y variables. The correlation coefficient "r" Correlation can only be used to describe quantitative variables. Categorical variables don’t have means and standard deviations.

9. Properties (cont.) r ranges from -1 to+1 "r" quantifies the strength and direction of a linear relationship between 2 quantitative variables. Strength: how closely the points follow a straight line. Direction: is positive when individuals with higher X values tend to have higher values of Y.

10. Properties (cont.) High correlation does not imply cause and effect CARROTS: Hidden terror in the produce department at your neighborhood grocery Everyone who ate carrots in 1920, if they are still alive, has severely wrinkled skin!!! Everyone who ate carrots in 1865 is now dead!!! 45 of 50 17 yr olds arrested in Raleigh for juvenile delinquency had eaten carrots in the 2 weeks prior to their arrest !!!

11. Properties (cont.) Cause and Effect There is a strong positive correlation between the monetary damage caused by structural fires and the number of firemen present at the fire. (More firemen-more damage) Improper training? Will no firemen present result in the least amount of damage?

12. Properties (cont.) Cause and Effect r measures the strength of the linear relationship between x and y; it does not indicate cause and effect correlation r =.935 x = fouls committed by player; y = points scored by same player (1,2) (24,75) (1,0) (18,59) (9,9) (3,7) (5,35) (20,46) (1,0) (3,2) (22,57) The correlation is due to a third “lurking” variable – playing time

13. 12 Simple Linear Regression Review 1. review of scatterplots and correlation 2. review of least squares procedure 3. inference for least squares lines

14. 13 House size House Cost Most lots sell for $25,000 Building a house costs about $75 per square foot. House cost = 25000 + 75(Size) The Model The model has a deterministic and a probabilistic components

15. 14 House cost = 25000 + 75(Size) House size House Cost Most lots sell for $25,000   However, house cost vary even among same size houses! The Model Since cost behave unpredictably, we add a random component.

16. 15 The Model The first order linear model y = dependent variable x = independent variable  0 = y-intercept  1 = slope of the line  = error variable x y 00 Run Rise   = Rise/Run  0 and  1 are unknown population parameters, therefore are estimated from the data.

17. 16 Estimating the Slope  1 and Intercept  0 The estimates are determined by –drawing a sample from the population of interest, –calculating sample statistics. –producing a straight line that cuts into the data.           Question: What should be considered a good line? x y

18. 17 The Least Squares (Regression) Line A good line is one that minimizes the sum of squared differences between the points and the line.

19. 18 The Least Squares (Regression) Line 3 3     4 1 1 4 (1,2) 2 2 (2,4) (3,1.5) Sum of squared differences =(2 - 1) 2 +(4 - 2) 2 +(1.5 - 3) 2 + (4,3.2) (3.2 - 4) 2 = 6.89 Sum of squared differences =(2 -2.5) 2 +(4 - 2.5) 2 +(1.5 - 2.5) 2 +(3.2 - 2.5) 2 = 3.99 2.5 Let us compare two lines The second line is horizontal The smaller the sum of squared differences the better the fit of the line to the data.

20. 19 The Estimated Coefficients To calculate the estimates of the slope and intercept of the least squares line, use the formulas: The least squares prediction equation that estimates the mean value of y for a particular value of x is:

21. 20 Example: –A car dealer wants to find the relationship between the odometer reading and the selling price of used cars. –A random sample of 100 cars is selected, and the data recorded. –Find the regression line. Independent variable x Dependent variable y The Simple Linear Regression Line

22. 21 The Simple Linear Regression Line Solution –Solving by hand: Calculate a number of statistics where n = 100.

23. 22 Solution – continued –Using the computer 1. Scatterplot 2. Trend function 3. Tools > Data Analysis > Regression The Simple Linear Regression Line

24. Regression Statistics Multiple R0.805167979 R Square0.648295475 Adjusted R Square0.644706653 Standard Error326.4886258 Observations100 ANOVA dfSSMSFSignificance F Regression119255607.37 180.6435.75078E-24 Residual9810446292.63106594.8228 Total9929701900 CoefficientsStandard Errort StatP-valueLower 95%Upper 95% Intercept17248.72734182.092574294.725045343.57E-9816887.3705617610.084 Odometer-0.066860890.004974639-13.440349285.75E-24-0.076732895-0.0569889 23 The Simple Linear Regression Line

25. 24 This is the slope of the line. For each additional mile on the odometer, the price decreases by an average of $0.0669 Interpreting the Linear Regression - Equation The intercept is b 0 = $17248.73. 0 No data Do not interpret the intercept as the “Price of cars that have not been driven” 17248.73

26. 25 Simple Linear Regression Review 1. review of scatterplots and correlation 2. review of least squares procedure 3. inference for least squares lines

27. 26 The Model The first order linear model y = dependent variable x = independent variable  0 = y-intercept  1 = slope of the line  = error variable x y 00 Run Rise   = Rise/Run  0 and  1 are unknown population parameters, therefore are estimated from the data.

28. 27 Error Variable: Required Conditions The error  is a critical part of the regression model. Four requirements involving the distribution of  must be satisfied. –The probability distribution of  is normal. –The mean of  is zero: E(  ) = 0. –The standard deviation of  is   for all values of x. –The set of errors associated with different values of y are all independent.

29. 28 The Normality of  From the first three assumptions we have: y is normally distributed with mean E(y) =  0 +  1 x, and a constant standard deviation   From the first three assumptions we have: y is normally distributed with mean E(y) =  0 +  1 x, and a constant standard deviation     0 +  1 x 1  0 +  1 x 2  0 +  1 x 3 E(y|x 2 ) E(y|x 3 ) x1x1 x2x2 x3x3  E(y|x 1 )  The standard deviation remains constant, but the mean value changes with x

30. 29 Assessing the Model The least squares method will produces a regression line whether or not there is a linear relationship between x and y. Consequently, it is important to assess how well the linear model fits the data. Several methods are used to assess the model. All are based on the sum of squares for errors, SSE.

31. 30 –This is the sum of differences between the points and the regression line. –It can serve as a measure of how well the line fits the data. SSE is defined by Sum of Squares for Errors –A shortcut formula

32. 31 –The mean error is equal to zero (recall: the mean of  is zero: E(  ) = 0). –If   is small the errors tend to be close to zero (close to the mean error). Then, the model fits the data well. –Therefore we can use   as a measure of the suitability of using a linear model. –An estimator of   is given by s  Estimate of    the Standard Deviation of the Error Term  s  is called the standard error since it is an estimate of the standard deviation  

33. 32 Example: –Calculate the standard error s  for the previous example and describe what it tells you about the model fit. Solution It is hard to assess the model based on s  even when compared with the mean value of y. Estimate of   an example

34. 33                       Testing the slope –When no linear relationship exists between two variables, the regression line should be horizontal.                                                                                           Different inputs (x) yield different outputs (y). No linear relationship. Different inputs (x) yield the same output (y). The slope is not equal to zeroThe slope is equal to zero Linear relationship.

35. 34 We can draw inference about  1 from b 1 by testing H 0 :  1 = 0 H 1 :  1 = 0 (or 0) –The test statistic is –If the error variable is normally distributed, the statistic is Student t distribution with d.f. = n-2. The standard error of b 1. where Testing the Slope

36. 35 Example –Test to determine whether there is enough evidence to infer that there is a linear relationship between the car auction price and the odometer reading for all three-year-old Tauruses in the previous example. Use  = 5%. Testing the Slope, Example

37. 36 Solving by hand –To compute “t” we need the values of b 1 and s b1. –The rejection region is t > t.025 or t < -t.025 with df = n-2 = 98, t.025 = 1.9845 Testing the Slope, Example

38. OdometerPrice 3740014600 4480014100Regression Statistics 4580014000Multiple R0.805167979 3090015600R Square0.648295475 3170015600 Adjusted R Square0.644706653 3400014700 Standard Error326.4886258 4590014500 Observation s100 1910015700 4010015100ANOVA 4020014800 dfSSMSFSignificance F 3240015200Regression119255607.3719255607.4180.6435.75078E-24 4350014700Residual9810446292.63106594.823 3270015600Total9929701900 3450015600 3770014600 CoefficientsStandard Errort StatP-valueLower 95%Upper 95% 4140014600Intercept17248.72734182.092574294.72504533.57E-9816887.3705617610.08 2450015700Odometer-0.0668608850.004974639-13.44034935.75E-24-0.076732895-0.05699 3580015000 4860014700 2420015400 37 Using the computer There is overwhelming evidence to infer that the odometer reading affects the auction selling price. Testing the Slope (Example)

39. Coefficient of determination 38 Reduction in prediction error when use x: TSS-SSE = SSR

40. Coefficient of determination 39 Reduction in prediction error when use x: TSS-SSE = SSR or TSS = SSR + SSE Overall variability in y TSS The regression model SSR Remains, in part, unexplained The error SSE Explained in part by

41. 40 Coefficient of determination: graphically x1x1 x2x2 y1y1 y2y2 y Two data points (x 1,y 1 ) and (x 2,y 2 ) of a certain sample are shown. Total variation in y = Variation explained by the regression line + Unexplained variation (error) Variation in y = SSR + SSE (TSS)

42. 41 Coefficient of determination R 2 (=r 2 ) measures the proportion of the variation in y that is explained by the variation in x. r 2 takes on any value between zero and one (-1  r  1). r 2 = 1: Perfect match between the line and the data points. r 2 = 0: There is no linear relationship between x and y.

43. 42 Example –Find the coefficient of determination for the used car price –odometer example. What does this statistic tell you about the model? Solution –Solving by hand; Coefficient of determination, Example

44. 43 – Using the computer From the regression output we have Coefficient of determination Regression Statistics Multiple R0.805167979 R Square0.648295475 Adjusted R Square0.644706653 Standard Error326.4886258 Observations100 ANOVA dfSSMSFSignificance F Regression119255607.37 180.6435.75078E-24 Residual9810446292.63106594.8228 Total9929701900 CoefficientsStandard Errort StatP-value Intercept17248.72734182.092574294.725045343.57E-98 Odometer-0.066860890.004974639-13.440349285.75E-24 64.8% of the variation in the auction selling price is explained by the variation in odometer reading. The rest (35.2%) remains unexplained by this model.

45. 44 If we are satisfied with how well the model fits the data, we can use it to predict the values of y. To make a prediction we use –Point prediction, and –Interval prediction Using the Regression Equation Before using the regression model, we need to assess how well it fits the data.

46. 45 Point Prediction Example –Predict the selling price of a three-year-old Taurus with 40,000 miles on the odometer. –It is predicted that a 40,000 miles car would sell for $14,574. – How close is this prediction to the real price? A point prediction

47. 46 Interval Estimates Two intervals can be used to discover how closely the predicted value will match the true value of y. –Prediction interval – predicts y for a given value of x, –Confidence interval – estimates the average y for a given x. –The confidence interval –The prediction interval

48. 47 Interval Estimates, Example Example - continued –Provide an interval estimate for the bidding price on a Ford Taurus with 40,000 miles on the odometer. –Two types of predictions are required: A prediction for a specific car An estimate for the average price per car

49. 48 Interval Estimates, Example Solution –A prediction interval provides the price estimate for a single car: t.025,98

50. 49 Solution – continued –A confidence interval provides the estimate of the mean price per car for a Ford Taurus with 40,000 miles reading on the odometer. The confidence interval (95%) = Interval Estimates, Example

51. 50 –As x moves away from x the interval becomes longer. That is, the shortest interval is found at x. The effect of the given x on the length of the interval

52. 51 –As x moves away from x the interval becomes longer. That is, the shortest interval is found at x = x. The effect of the given x on the length of the interval

53. 52 –As x moves away from x the interval becomes longer. That is, the shortest interval is found at x = x. The effect of the given x on the length of the interval

54. 53 Regression Diagnostics - I The three conditions required for the validity of the regression analysis are: –the error variable is normally distributed. –the error variance is constant for all values of x. –The errors are independent of each other. How can we diagnose violations of these conditions?

55. 54 Residual Analysis Examining the residuals (or standardized residuals), help detect violations of the required conditions. Example – continued: –Nonnormality. Use Excel to obtain the standardized residual histogram. Examine the histogram and look for a bell shaped. diagram with a mean close to zero.

56. 55 For each residual we calculate the standard deviation as follows: A Partial list of Standard residuals Standardized residual ‘i’ = Residual ‘i’ Standard deviation Residual Analysis

57. 56 It seems the residual are normally distributed with mean zero Residual Analysis

58. 57 Heteroscedasticity When the requirement of a constant variance is violated we have a condition of heteroscedasticity. Diagnose heteroscedasticity by plotting the residual against the predicted y. + + + + + + + + + + + + + + + + + + + + + + + + The spread increases with y ^ y ^ Residual ^ y + + + + + + + + + + + + + + + + + + + + + + +

59. 58 Homoscedasticity When the requirement of a constant variance is not violated we have a condition of homoscedasticity. Example - continued

60. 59 Non Independence of Error Variables – A time series is constituted if data were collected over time. –Examining the residuals over time, no pattern should be observed if the errors are independent. –When a pattern is detected, the errors are said to be autocorrelated. –Autocorrelation can be detected by graphing the residuals against time.

61. 60 Patterns in the appearance of the residuals over time indicates that autocorrelation exists. + + + + + + + + + + + + + + + + + + + + + + + + + Time Residual Time + + + Note the runs of positive residuals, replaced by runs of negative residuals Note the oscillating behavior of the residuals around zero. 00 Non Independence of Error Variables

62. 61 Outliers An outlier is an observation that is unusually small or large. Several possibilities need to be investigated when an outlier is observed: –There was an error in recording the value. –The point does not belong in the sample. –The observation is valid. Identify outliers from the scatter diagram. It is customary to suspect an observation is an outlier if its | standard residual | > 2

63. 62 + + + + + + + + + + + + + + + + + The outlier causes a shift in the regression line … but, some outliers may be very influential ++++++++++ An outlier An influential observation

64. 63 Procedure for Regression Diagnostics Develop a model that has a theoretical basis. Gather data for the two variables in the model. Draw the scatter diagram to determine whether a linear model appears to be appropriate. Determine the regression equation. Check the required conditions for the errors. Check the existence of outliers and influential observations Assess the model fit. If the model fits the data, use the regression equation.