Docsity
Docsity

Prepare for your exams
Prepare for your exams

Study with the several resources on Docsity


Earn points to download
Earn points to download

Earn points by helping other students or get them with a premium plan


Guidelines and tips
Guidelines and tips

Relative Standard Deviation: Computing and Understanding this Statistical Measure, Study notes of Statistics

The concept of relative standard deviation (rsd), its purpose, syntax, and differences with other statistical measures such as coefficient of variation. It also provides examples of how to use the relative standard deviation let subcommand in sas and its applications in exploratory data analysis. The document also includes a reference to a statistical methods book and an implementation date.

Typology: Study notes

2021/2022

Uploaded on 09/12/2022

freddye
freddye 🇺🇸

4.3

(11)

235 documents

1 / 2

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
RELATIVE STANDARD DEVIATION Statistics LET Subcommands
2-46 September 3, 1996 DATAPLOT Reference Manual
RELATIVE STANDARD DEVIATION
PURPOSE
Compute the relative standard deviation of a variable.
DESCRIPTION
The relative standard deviation is:
relsd = 100*(standard deviation/|mean|)
Some analysts prefer to call this the percent relative standard deviation and to call the relative standard deviation the value that is
obtained without multiplying by 100. If this is your preference, simply use the LET command to divide by 100.
Note that this statistic differs in a minor way from the commonly used coefficient of variation. The coefficient of variation divides by
the mean rather than the absolute value of the mean. If the mean is negative, the coefficient of variation will be negative while the
relative standard deviation (as defined here) will always be positive. In addition, some sources define the coefficient of variation as a
fraction rather than a percent. To calculate the traditional coefficient of variation, use the COEFFICIENT OF VARIATION LET
subcommand.
SYNTAX
LET <par> = RELATIVE STANDARD DEVIATION <y> <SUBSET/EXCEPT/FOR qualification>
where <y> is a response variable;
<par> is a parameter where the relative standard deviation value is stored;
and where the <SUBSET/EXCEPT/FOR qualification> is optional.
EXAMPLES
LET RSD = RELATIVE STANDARD DEVIATION Y1
LET RSD = RELATIVE STANDARD DEVIATION Y1 SUBSET TAG > 2
LET PERRSD = RELATIVE STANDARD DEVIATION Y1
LET RSD = PERRSD/100
NOTE
Versions prior to 94/2 divide by the mean rather than the absolute value of the mean. The 94/12 version implements the COEFFICIENT
OF VARIATION command as a distinct command. The COEFFICIENT OF VARIATION command divides by the mean rather than the
absolute value of the mean.
DEFAULT
None
SYNONYMS
RELATIVE SD
RELATED COMMANDS
COEFFICIENT OF VARIATION = Compute the coefficient of variation of a variable.
MEAN = Compute the mean of a variable.
STANDARD DEVIATION = Compute the standard deviation of a variable.
RELATIVE VARIANCE = Compute the relative variance of a variable.
RELATIVE STAND DEVI PLOT = Generate a relative standard deviation versus subset plot.
REFERENCE
“Statistical Methods,” Sixth Edition, Snedecor and Cochran, The Iowa State University Press, 1967 (pp. 62-64).
APPLICATIONS
Exploratory Data Analysis
IMPLEMENTATION DATE
Pre-1987 (definition changed 94/2)
pf2

Partial preview of the text

Download Relative Standard Deviation: Computing and Understanding this Statistical Measure and more Study notes Statistics in PDF only on Docsity!

RELATIVE STANDARD DEVIATION Statistics LET Subcommands

2-46 September 3, 1996 DATAPLOT Reference Manual

RELATIVE STANDARD DEVIATION

PURPOSE

Compute the relative standard deviation of a variable.

DESCRIPTION

The relative standard deviation is:

relsd = 100*(standard deviation/|mean|)

Some analysts prefer to call this the percent relative standard deviation and to call the relative standard deviation the value that is obtained without multiplying by 100. If this is your preference, simply use the LET command to divide by 100.

Note that this statistic differs in a minor way from the commonly used coefficient of variation. The coefficient of variation divides by the mean rather than the absolute value of the mean. If the mean is negative, the coefficient of variation will be negative while the relative standard deviation (as defined here) will always be positive. In addition, some sources define the coefficient of variation as a fraction rather than a percent. To calculate the traditional coefficient of variation, use the COEFFICIENT OF VARIATION LET subcommand.

SYNTAX

LET = RELATIVE STANDARD DEVIATION <SUBSET/EXCEPT/FOR qualification> where is a response variable; is a parameter where the relative standard deviation value is stored; and where the <SUBSET/EXCEPT/FOR qualification> is optional.

EXAMPLES

LET RSD = RELATIVE STANDARD DEVIATION Y

LET RSD = RELATIVE STANDARD DEVIATION Y1 SUBSET TAG > 2

LET PERRSD = RELATIVE STANDARD DEVIATION Y

LET RSD = PERRSD/

NOTE

Versions prior to 94/2 divide by the mean rather than the absolute value of the mean. The 94/12 version implements the COEFFICIENT OF VARIATION command as a distinct command. The COEFFICIENT OF VARIATION command divides by the mean rather than the absolute value of the mean.

DEFAULT

None

SYNONYMS

RELATIVE SD

RELATED COMMANDS

COEFFICIENT OF VARIATION = Compute the coefficient of variation of a variable. MEAN = Compute the mean of a variable. STANDARD DEVIATION = Compute the standard deviation of a variable. RELATIVE VARIANCE = Compute the relative variance of a variable. RELATIVE STAND DEVI PLOT = Generate a relative standard deviation versus subset plot.

REFERENCE

“Statistical Methods,” Sixth Edition, Snedecor and Cochran, The Iowa State University Press, 1967 (pp. 62-64).

APPLICATIONS

Exploratory Data Analysis

IMPLEMENTATION DATE

Pre-1987 (definition changed 94/2)

Statistics LET Subcommands RELATIVE STANDARD DEVIATION

DATAPLOT Reference Manual September 3, 1996 2-

PROGRAM

SKIP 25

READ GEAR.DAT DIAMETER

LET A = RELATIVE STANDARD DEVIATION DIAMETER

The computed relative standard deviation is 0.63.