企业管理研究方法（英文版·第11版）唐纳德·库珀-Chap018.pptVIP

* Exhibit 18-16 A residual is what remains after the line is fit. When standardized, residuals are comparable to Z scores with a mean of 0 and a standard deviation of 1. In this plot, the standardized residuals should fall between 2 and -2, be randomly distributed about zero, and show no discernible pattern. All these conditions say the model is applied appropriately. * Exhibit 18-17 Prediction and confidence bands are bow-tie shaped confidence interval around a predictor. Predictors farther from the mean have larger bandwidths. If we wanted to predict the price of a case of investment-grade red wine for a growing season that averages 21 degrees Celsius, our prediction would be Use formula from page 552 This is a point prediction Y and should be corrected for greater precision. As with other confidence estimates, we establish the degree of confidence desired and substitute into the formula Use formula from page 552 * With the regression line plotted, next we evaluate the goodness of fit. Goodness of fit is a measure of how well the regression model is able to predict Y. The most important test in bivariate linear regression is whether the slope, ?1 , is equal to zero. Zero slopes result from the conditions listed in the slide. To test whether the slope is equal to zero, we use a two-tailed test. The test follows the t distribution for n-2 degrees of freedom. In this case, we reject the null hypothesis because the calculated t is greater than any t value for 8 degrees of freedom and an alpha of .01. * Exhibit 18-18 Computer printouts generally contain an analysis of variance (ANOVA) table with an F test of the regression model. In bivariate regression, t and F tests produce the same results since t2 is equal to F. In multiple regression, the F test has an overall role for the model and each of the independent variables is evaluated in a separate t-test. For regression, the ANOVA comprises explained deviations and unexplained deviations. Together, the