/* SAS Example 2: Textbook Chapter 13 Question 13.5 */ /* Fortune Magazine published data on the worlds largest industrial corporations. Information was given on sales, profits, assets, stockholders' equity, and number of employees. Listed here are the data for a sample of 12 of these companies for four variables. Use the data to determine the equation of the multiple regression model to predict profits by revenue, assets and number of employees. Revenue, profits and assets are given in millions of dollars. */ /* The OPTIONS NODATE command will omit the date from our output if we want to make it look tidier. */ OPTIONS PS=52; OPTIONS LS=78; OPTIONS NODATE; OPTIONS PAGENO=1; DATA FORTUNE; INPUT COMPANY $ 1-30 PROFITS ASSETS EMPLOYEES REVENUES; CARDS; Nestle (Switzerland) 2377.8 34562.6 212687 41625.7 Hitachi (Japan) 1146.7 105257.5 331673 76430.9 Renault (France) 655.5 42357.4 138279 32188.0 Rhone-Poulenc (France) 547.7 22947.1 81582 15559.5 Volvo (Sweden) 1714.9 18652.9 60115 20204.0 Petrofina (Belgium) 324.9 10868.2 14013 11399.1 Dow (United States) 931.0 26545.0 53700 20015.0 Pohang (South Korea) 474.9 15676.7 22891 9064.1 Indian Oil (India) 326.6 5807.3 34729 8235.7 Sandoz (Switzerland) 1268.7 14935.1 60304 11611.1 Raytheon (United States) 596.9 7395.4 60200 10012.9 Akzo Nobel (Netherlands) 647.5 10364.1 70400 12206.9 ; RUN; /* The following command will print out the data set, and the title command can be inserted into any proc statement if we want to put a heading over our output */ PROC PRINT DATA=FORTUNE; TITLE 'Industrial Corporation Data'; RUN; /* If we wish to examine the correllation between our variables, we can use the proc corr command */ PROC CORR; VAR PROFITS ASSETS EMPLOYEES REVENUES; RUN; /* The CLM statement asks for confidence intervals for the mean, and CLI asks for prediction intervals for the individual */ PROC REG; TITLE 'Model To Predict Profits with Assets, Number of Employees and Revenue'; MODEL PROFITS = ASSETS EMPLOYEES REVENUES/ P R CLM CLI; RUN;