Compute the ratio of a user specified statistic for two response
variables.
Description:
When comparing the scale statistics of two response variables, it
is common to compare the ratio of the statistics rather than the
difference. For example, given the variables Y1 and Y2, the command
LET R = RATIO OF STANDARD DEVIATION Y1 Y2
is equivalent to entering
LET S1 = STANDARD DEVIATION Y1
LET S2 = STANDARD DEVIATION Y2
LET R = S1/S2
The advantage of using the RATIO OF command is that you can use
the ratio statistic in commands such as the
BOOTSTRAP PLOT and
STATISTIC PLOT.
The statistic specified in the RATIO OF command must be one that
is computed for a single response variable (e.g., the CORRELATION
statistic is not supported by this command). Although the RATIO OF
command was motivated by scale statistics, it can be used for any
supported statistic for a single response variable.
Syntax:
LET <par> = RATIO OF <stat> <y>
<SUBSET/EXCEPT/FOR qualification>
where <stat> is one of Dataplot's supported statistics for a single
response variable;
<y> is the response variable;
<par> is a parameter where the computed ratio is saved;
and where the <SUBSET/EXCEPT/FOR qualification> is optional.
Examples:
LET A = RATIO OF STANDARD DEVIATIONS Y1 Y2
LET A = RATIO OF STANDARD DEVIATIONS Y1 Y2 SUBSET Y2 > 0
LET A = RATIO OF MEDIAN ABSOLUTE DEVIATIONS Y1 Y2
LET A = RATIO OF RANGE Y1 Y2
LET A = RATIO OF INTERQUARTILE RANGE Y1 Y2
Note:
Dataplot statistics can be used in a number of commands. For
details, enter
