商务统计-C14商务统计.pdf

Chapter 14 Mult iple Linear Regression and Corre lat ion Ana lysis GOALS 1.Describe the re lat ionship between several independent variables and a dependent variable using mult ip le regression analysis. 2.Set up, interpret, and apply an ANOVA table 3.Compute and interpret the mult ip le standard error of est imate, the coeffic ient of mult ip le determ inat ion, and the adjusted coeffic ient of mult ip le determ inat ion. GOALS 4.Conduct a test of hypothesis to determ ine whether regression coeffic ients d iffer from zero. 5.Conduct a test of hypothesis on each of the regression coeffic ients. 6.Use residual analysis to evaluate the assumpt ions of mult iple regression analysis. 7.Evaluate the effects of corre lated independent variables. GOALS 8.Use and understand qualitat ive independent variables. Mult iple Regression Analysis The general mult iple regression w ith k independent variables is given by: The least squares criterion is used to deve lop this equat ion. Because determ ining b , b , etc. is 1 2 very ted ious, a software package such as Exce l or MINITAB is recommended. Mult iple Regression Analysis For two independent variables, the general form of the mult iple regression equat ion is: Mult iple Regression Analysis • X and X are the independent variables. 1 2 • a is the Y- intercept • b is the net change in Y for each unit change in 1 X ho ld ing X constant. It is called a part ial 1 2 regression coeffic ient, a net regression coeffic ient, or just a regression coeffic ient. Regression Plane for a 2- Independent Variable Linear Regression Equat ion Mult iple Linear Regression - Example Salsberry Realty se lls homes along the east coast of the United States. One of the quest ions most frequent ly asked by prospect ive buyers