home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Magnum One
/
Magnum One (Mid-American Digital) (Disc Manufacturing).iso
/
d13
/
stistics.arc
/
MULTREGR.DOC
< prev
next >
Wrap
Text File
|
1988-07-24
|
3KB
|
57 lines
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)