How to fit a (left) skewed distribution

Hi,

Currently, I am using the following code to fit distributions to my data:

pd = fitdist(mydata,distribution)
x_values = 1:1:26;
y = pdf(pd,x_values);
plot(x_values,y,'LineWidth',2)

However, as far as I can see all the distributions offered by Matlab are either not skewed or right-skewed. Hence my question: Knowing that my data is left-skewed, how could I fit such a distribution to it?

Thanks

[Kin Sung Chan]

histfit(mydata/max(mydata), 30, 'beta'); % let's say using bin = 30.

However, even though it can produce a fit, the 'beta' requires your data to be within 0 and 1.

Answer 1

[Image Analyst]

Try fitting the max value minus your data. In other words, flip your distribution right to left.

[MiauMiau]

So you mean to change my distribution to a right-skewed distribution? And then how would I proceed..? Showing a right skewed distribution for an underlying left-skewed dataset is, I assume, quite confusing for the audience.

[Image Analyst]

If you attach mydata in a .mat file, maybe someone will show you.

[MiauMiau]

That would be very nice:) It is attached

[Image Analyst]

It looks fine to me

% Initialization steps.
% clc;    % Clear the command window.
close all;  % Close all figures (except those of imtool.)
clear;  % Erase all existing variables. Or clearvars if you want.
workspace;  % Make sure the workspace panel is showing.
format long g;
format compact;
fontSize = 20;

storedStructure = load('mydata.mat')
mydata = storedStructure.mydata';
subplot(2, 1, 1);
plot(mydata)
grid on;
title('Original Data', 'FontSize', fontSize);
subplot(2, 1, 2);
hObj = histogram(mydata, 10)
grid on;
title('Histogram of Original Data', 'FontSize', fontSize);

distribution = 'Lognormal';
pd = fitdist(mydata,distribution)
x_values = linspace(0, max(mydata), 100);
y = pdf(pd,x_values);
% subplot(2, 2, 3);
hold on;
plot(x_values,y,'LineWidth',2)
title('Original Data', 'FontSize', fontSize);
legend('Data', 'Fitted Distribution');

% Set up figure properties:
% Enlarge figure to full screen.
set(gcf, 'Units', 'Normalized', 'OuterPosition', [0, 0.04, 1, 0.96]);
% Get rid of tool bar and pulldown menus that are along top of figure.
% set(gcf, 'Toolbar', 'none', 'Menu', 'none');
% Give a name to the title bar.
set(gcf, 'Name', 'Demo by ImageAnalyst', 'NumberTitle', 'Off')

[Maria Amr]

Image Analyst, MiauMiau

I have the same problem but my data are right skewed. Would you please direct me how to fit a right skewed distributions? Thank you!

[Image Analyst]

Attach your data in a new question. In the meantime, see attached demos and adapt as needed.