Is it possible to change the order of groups in the GUI created by the MULTCOMPARE function in the Statistics Toolbox?

Question

MathWorks Support Team 2009-6-27

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/101288-is-it-possible-to-change-the-order-of-groups-in-the-gui-created-by-the-multcompare-function-in-the-s

I am using the Statistics Toolbox function MULTCOMPARE function with the "stats" structure created by ANOVAN. I want to change the order of the groups on the Y-axis of the MULTCOMPARE GUI. For example:

y = [52.7 57.5 45.9 44.5 53.0 57.0 45.9 44.0]';
g1 = [1 2 3 2 1 2 1 2]; 
g2 = {'hi';'hi';'lo';'lo';'hi';'hi';'lo';'lo'}; 
g3 = {'may'; 'may'; 'may'; 'may'; 'june'; 'june'; 'june'; 'june'};
[p,table,stats] = anovan(y, {g1 g2 g3});
multcompare(stats);

I would like the labels and corresponding data for "X1=1" and "X1=3" to be interchanged. I would like MULTCOMPARE to accept an input that allows for control of the ordering of the Y-axis.

请先登录，再回答此问题。

Answer 1

MathWorks Support Team 2009-6-27

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/101288-is-it-possible-to-change-the-order-of-groups-in-the-gui-created-by-the-multcompare-function-in-the-s#answer_110636

The ability to place the Y-axis data of the MULTCOMPARE GUI in an arbitrary user specified order via an input argument is not available in the Statistics Toolbox.

To work around this issue, it is possible to reverse the ordering of the Y-axis data by setting the "YDir" property of the MULTCOMPARE axes to "reverse"

y = [52.7 57.5 45.9 44.5 53.0 57.0 45.9 44.0]';
g1 = [1 2 3 2 1 2 1 2]; 
g2 = {'hi';'hi';'lo';'lo';'hi';'hi';'lo';'lo'}; 
g3 = {'may'; 'may'; 'may'; 'may'; 'june'; 'june'; 'june'; 'june'};
[p,table,stats] = anovan(y, {g1 g2 g3});
multcompare(stats);
set(gca,'YDir','Reverse');

For input data from ANOVA1, ANOVA2, and ANOVAN, it may also be possible to achieve a desired ordering on the Y-axis of MULTCOMPARE by properly manipulating the order of the "stats" structure returned by a particular test. For example:

      [p,t,st] = anova1(MPG,Origin,'off');
      st_new  = st;
      st_new.gnames(1) = st_new.gnames(6);
      st_new.n(1) = st_new.n(6);
      st_new.means(1) = st_new.means(6);
      st_new.gnames(6) = st.gnames(1);
      st_new.n(6) = st.n(1);
      st_new.means(6) = st.means(1);
      [c,m,h,nms] = multcompare(st_new)