home *** CD-ROM | disk | FTP | other *** search
- MULTIPLE LINEAR REGRESSION
-
- The program will do multiple linear regression, for example, on age, weight
- and height of 8 boys, using weight as the dependent variable and age/height
- as independent variables. This would permit one to estimate the weight of
- a 7 year old boy who is 51 inches tall.
-
- EXAMPLE: INDEPENDENT INDEPENDENT
- VARIABLE VARIABLE DEPENDENT
- Here is sample ____#1_________#2________VARIABLE
- data for 10 boys: AGE HEIGHT WEIGHT
- years inches pounds
-
- 8 59 48
- 9 55 49
- 6 50 44
- 10 80 59
- 8 61 55
- 9 75 51
- 9 67 55
- 7 58 50
-
- After entering this raw data, the program will print equation coefficients,
- coefficient of determination, coefficient of multiple correlation and the
- standard error of the estimate. It will then ask for interpolation input
- (for example, you might want to know the weight of a 7 year old whose
- height is 51 inches; you would then enter 7 as variable one and 51 as
- variable 2). It will then calculate that his weight would be 46.86 pounds.
-
- The raw data may be entered via a data file, created in the NON-DOCUMENT
- mode of WordStar. The format for this file is shown below. Data for
- interpolation must be entered from the keyboard, once the initial raw data
- is entered.
-
- MULTIPLE REGRESSION ANALYSIS SAMPLE DATA FILE:
-
- Multiregression test data: <-- Title for the run, up to 65 characters.
- 8 <-- Number of data points (rows).
- 2 <-- Number of independent variables.
- 8 59 48 <-- Data (compare with table above).
- 9 55 49 .
- 6 50 44 .
- 10 80 59 .
- 8 61 55 .
- 9 75 51 .
- 9 67 55 .
- 7 58 50 .
-
-
- REFERENCE: Poole, L. and M. Borchers 1979 Some Common Basic Programs,
- 3rd edition. Osborne/McGraw-Hill, Berkeley, CA. Page 147-150.
-
- BASIC VERSION: Poole and Borchers coded Mar 84
- PASCAL VERSION: Dr. Stanley Kaplan Sep 85
- UPDATES: Oct 86 (Version 2.00)
- Feb 87 (Version 2.10)
- Feb 88 (MS-DOS Version 2.2)