visualdatatools.com Forum Index visualdatatools.com
Discussion for DataTank and DataGraph
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Measure of the width of a distribution?

 
Post new topic   Reply to topic    visualdatatools.com Forum Index -> Request Fit Function
View previous topic :: View next topic  
Author Message
paula



Joined: 14 Jan 2013
Posts: 3

PostPosted: Thu Jan 24, 2013 12:33 pm    Post subject: Measure of the width of a distribution? Reply with quote

I apologize for the second posting of this same topic, but I realized that this might be a more appropriate place for it than the "New Feature Request".
Thanks for any feedback

Greetings,

I have a bit of an unusual question, but I'm wondering if anyone has more experience with distribution modeling than I do. I do research on organic semiconductor thin films, and when I scan samples, I create a histogram of energies at every site on the surface.

For most samples, the histogram/energy distribution starts out very asymmetric then progresses to a more Gaussian distribution. So, what I'm looking for is essentially a measure of "Gaussian-ness" so that I can give a quantitative measure of how asymmetric a sample starts out compared to the asymmetry when I'm finished scanning.

I have a feeling I'm looking for something like a chi-squared measure, and I've also had a "moment" analysis suggested. Has anyone used the chi-squared measure within data graph? Also, does anyone have a suggestion of a reference to explain what a "good" chi-squared value is for a Gaussian distribution?

Thanks!
Back to top
View user's profile Send private message
Mike



Joined: 05 May 2008
Posts: 277
Location: Christchurch, New Zealand

PostPosted: Mon Jan 28, 2013 10:59 pm    Post subject: Reply with quote

I don't have anything useful to say about the chi square measure here. But have you looked at the skewness and kurtosis measures that are provided by the histogram command?

I think a perfect gaussian distribution should have a skewness of zero and kurtosis of 3, so that gives you some point of reference to use, but not sure how to take that further to an inferential test.
Back to top
View user's profile Send private message Visit poster's website
paula



Joined: 14 Jan 2013
Posts: 3

PostPosted: Wed Jan 30, 2013 1:27 pm    Post subject: Reply with quote

That actually helps quite a bit. I'd had someone suggest a moment analysis, but couldn't find the right method of execution. Your calling it a "skewness" helped me find it.

Thanks so much!
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    visualdatatools.com Forum Index -> Request Fit Function All times are GMT - 3 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group