Dataplot Vol 2 Vol 1

# WEIGHTED SUM OF SQUARES

Name:
WEIGHTED SUM OF SQUARES (LET)
Type:
Let Subcommand
Purpose:
Compute the weighted sum of squares of a variable.
Description:
The weighted sum of squares is defined as

where X is the response variable and W is the weights variable. The response variable and weights variable must have the same number of observations.

For this command, the weights are not normalized. However, at least one of the weights must be positive and none of the weights can be negative. Otherwise, an error message is reported.

Syntax:
LET <par> = WEIGHTED SUM OF SQUARES<x> <w>
<SUBSET/EXCEPT/FOR qualification>
where <x> is the response variable;
<w> is the weights variable;
<par> is a parameter where the weighted sum of squares is saved;
and where the <SUBSET/EXCEPT/FOR qualification> is optional.
Examples:
LET A = WEIGHTED SUM OF SQUARES Y1 WEIGHT
LET A = WEIGHTED SUM OF SQUARES Y1 WEIGHT SUBSET Y1 > 0
Note:
Dataplot statistics can be used in a number of commands. For details, enter

Default:
None
Synonyms:
None
Related Commands:
 WEIGHTED SUM = Compute the weighted sum of a variable. WEIGHTED SUM OF ABSOLUTE VALUES = Compute the weighted sum of absolute values of a variable. WEIGHTED AVERAGE OF ABSOLUTE VALUES = Compute the weighted average of absolute values of a variable. WEIGHTED SUM OF DEVIATIONS FROM THE MEAN = Compute the weighted sum of deviations from the mean of a variable. WEIGHTED SUM OF SQUARED DEVIATIONS FROM MEAN = Compute the weighted sum of squared deviations from the mean of a variable. WEIGHTED MEAN = Compute the weighted mean of a variable. WEIGHTED STANDARD DEVIATION = Compute the weighted standard deviation of a variable. WEIGHTED VARIANCE = Compute the weighted variance of a variable.
Applications:
Data Analysis
Implementation Date:
2012/06
Program:
```
LET Y = DATA 2 3 5 7 11 13 17 19 23
LET W = DATA 1 1 0 0 4 1 2 1 0
.
LET A = WEIGHTED SUM OF SQUARES Y W
```
The returned value is 1605.

Date created: 06/29/2012
Last updated: 06/29/2012
Please email comments on this WWW page to alan.heckert@nist.gov.