Doeeffectsplot: Difference between revisions
Jump to navigation
Jump to search
imported>Jeremy No edit summary |
imported>Jeremy (→Inputs) |
||
Line 10: | Line 10: | ||
*'''doe''' = DOE dataset object. | *'''doe''' = DOE dataset object. | ||
*'''icol''' = user selected doe column index. DOE columns are arranged as factors first, then interactions. | *'''icol''' = user selected doe column index. DOE columns are arranged as factors first, then interactions. | ||
:For example: <tt>F1, F2, F3, F1xF2, F1xF3, F2xF3</tt> | *:For example: <tt>F1, F2, F3, F1xF2, F1xF3, F2xF3</tt> | ||
*'''alpha''' = 0.05 for two-sided critical region. | *'''alpha''' = 0.05 for two-sided critical region. | ||
*'''y''' = experimentally measured response values | *'''y''' = experimentally measured response values | ||
====Optional Inputs==== | ====Optional Inputs==== | ||
*'''options''' = structure array with the following fields: | *'''options''' = structure array with the following fields: |
Latest revision as of 13:08, 22 February 2013
Purpose
Create main effect or interaction plot, incl LSD bars.
Synopsis
- doeeffectsplot(doe, y, icol, alpha)
- doeeffectsplot(doe, y, icol, alpha, options)
Description
Main & Interaction Effects plots which contain Fisher's Least Significant Difference bars around the mean effect values.
Inputs
- doe = DOE dataset object.
- icol = user selected doe column index. DOE columns are arranged as factors first, then interactions.
- For example: F1, F2, F3, F1xF2, F1xF3, F2xF3
- alpha = 0.05 for two-sided critical region.
- y = experimentally measured response values
Optional Inputs
- options = structure array with the following fields:
- display: [ 'off' | {'on'} ] governs level of display to command window.
Example
Use doegui or doegen to create a doe object, for example with 3 factors (with 2 or more levels) and including 2 term interactions. Then run the following to set up:
icol=1; alpha = 0.05; y = rand(size(doe,1),1); % create y, some expt result values [doe.data(:,icol) y] % view Factor 1 and y values %Add the following to make significant difference between F1 groups y = y + doe.data(:,icol); doeeffectsplot(doe, y, icol, alpha)