How to modify Box Plot outliers marker size?? Can I insert a legend?

Claudio Ignacio Fernandez (view profile)

on 3 Jul 2019
Latest activity Commented on by Claudio Ignacio Fernandez

on 3 Jul 2019

the cyclist (view profile)

Hello community
I'm working with box plot, I built them using previous post related to the subject.
My main 2 problems here are:
How can I increase the size of the markers for outliers? Because when I'm doing my report is very difficult to see wich one in 'o' or 'x'.
Also, based in this code, how can I add a legend "group 1", "group 2" ?
Here is the code I'm using
clc
clear
var =file1;
x1 = var(1:124,1);
x2 = var(1:124,2);
x3 = var(1:124,3);
x4 = var(1:124,4);
x5 = var(1:124,5);
x6 = var(1:124,6);
x7 = var(1:124,7);
x11 = var(125:225,1);
x12 = var(125:225,2);
x13 = var(125:225,3);
x14 = var(125:225,4);
x15 = var(125:225,5);
x16 = var(125:225,6);
x17 = var(125:225,7);
x = [x1;x2;x3;x4;x5;x6;x7];
xx = [x11;x12;x13;x14;x15;x16; x17];
group_1 = [ones(size(x1)); 2*ones(size(x2)); 3*ones(size(x3));4*ones(size(x4)); 5*ones(size(x5)); 6*ones(size(x6)); 7*ones(size(x7))];
group_2 = [ones(size(x11)); 2*ones(size(x12)); 3*ones(size(x13));4*ones(size(x14)); 5*ones(size(x15)); 6*ones(size(x16)); 7*ones(size(x17))];
position1 = [2.325 3.325 4.325 5.325 6.325 7.325 8.325 ];
position2 = [2.575 3.575 4.575 5.575 6.575 7.575 8.575];
figure()
boxplot(x,group_1,'color','g','Widths',0.2,'positions',position1,'Symbol','o');
hold on
grid on
set(gca, 'fontweight', 'bold', 'box', 'on', 'fontsize', 18);
boxplot(xx,group_2,'color','r','Widths',0.2,'positions',position2, 'Symbol','x','Labels',{'0','1','2','3','4','5','6'});
set(findobj(gca,'type','line'),'linew',1.5)
xlabel('Days from Inoculation')
ylabel('Scores PCA 1')
ax = gca;
ax.XRuler.Axle.LineWidth = 2;
ay = gca;
ay.YRuler.Axle.LineWidth = 2;
hold off

the cyclist

the cyclist (view profile)

on 3 Jul 2019
I formatted your code using the CODE section. Please use that next time.
Would you be able to upload your file1, so that we can run your code and see exactly what you are seeing? That is sometimes more helpful than answering in the abstract.
Claudio Ignacio Fernandez

on 3 Jul 2019
Thank you.

R2019a

the cyclist (view profile)

on 3 Jul 2019
Edited by the cyclist

the cyclist (view profile)

on 3 Jul 2019

I think the following should work for you:
figure
boxplot(randn(1000,2))
h = findobj(gcf,'tag','Outliers')
set(ho,'MarkerSize',24)
Where I have written gcf, you should put your figure. (I was just using the current figure.)
I made the outliers huge, so the change is obvious. You can adjust the size with that parameter I set to 24.
You might have to adjust exactly how you do all this, given different number of boxes, outliers, etc, but you should get the gist.

1 Comment

Claudio Ignacio Fernandez

on 3 Jul 2019