How to identify the distribution of the data set?
108 views (last 30 days)
Show older comments
Given a data set (.mat file). How can I identify the distribution (Normal, Gaussian, etc) of the data in matlab? Is there any built-in function that helps to do this?
For example, I'd like to identify the distribution of the Ionosphere data set.
0 Comments
Accepted Answer
John
on 1 Jul 2016
Plot the histogram. Use 'hist'. Depending on the number of samples in the data set, you will want to increase the number of bins accordingly to get better resolution. As with all statistical analysis, the more data you sample, the better your fit will be.
1 Comment
alex brown
on 29 Apr 2019
Doubts about distribution of the answers generated by randfixedsum!
check this:
clc
close all
for i=1:1000
an(:,i)=randfixedsum(24,1,3600,5,300);
end
% histogram(an(1,:))
hold on
histogram(an(5,:))
look carefully at the distribution. I don't think it is uniformly distributed.
More Answers (2)
Image Analyst
on 3 Jul 2016
Of course the histogram is the actual distribution. But if you want a model distribution, look into kstest(), kstest2(), lillietest(), etc.
1 Comment
alex brown
on 29 Apr 2019
Thank you for helping me. Is there any function for testing the uniform distribution?! what is the criteria for a distribution to be uniform?!
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!