Fitting Normal Distribution Histogram - Help Needed

Click For Summary

Discussion Overview

The discussion revolves around fitting a histogram of a data set to a normal distribution, with participants exploring the suitability of different distribution models based on the characteristics of the data.

Discussion Character

  • Exploratory, Debate/contested, Technical explanation

Main Points Raised

  • One participant expresses difficulty in fitting their data to a normal distribution and questions whether they are making an error in their approach.
  • Another participant suggests that the data may not be normally distributed and proposes the possibility of it being bimodal.
  • A different participant mentions achieving better fitting results using an online program and shares the specific data points used, questioning why their results differ from those obtained online.
  • One participant agrees that the data should not be considered normally distributed and emphasizes that the sample size is too small to reliably determine the shape of the distribution, recommending alternative graphing methods.
  • A participant proposes the idea of using a skew normal distribution as a potential alternative fit for the data.

Areas of Agreement / Disagreement

Participants generally agree that the data does not appear to fit a normal distribution, but there is no consensus on the best alternative model or the implications of the sample size.

Contextual Notes

Participants note limitations related to sample size and the potential for different distribution shapes, but these aspects remain unresolved.

randa177
Messages
90
Reaction score
1
I have a set of data that I used to create a histogram, and would like to fit it with a Gauss shape...I used the regular equation for a normal distribution, using the mean value and the standard deviation of my data, but the Gauss shape doesn't seem to be really fitting my data... I might be doing something wrong... I am attaching it here...

Thanks!
 

Attachments

Physics news on Phys.org
Yeah, the data doesn't look normal. Maybe it's a bimodal distribution of some sort?
 
actually when I use the same of data on an online program I get better results...
These are the numbers I am using:
-2.1
-1.8
-1.6
-1.4
-1.2
-1.1
-1
-0.9
-0.8
-0.7
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.8

Do you know what the problem might be?
PS: Try putting these numbers here: http://azzalini.stat.unipd.it/SN/sn-fit.html
It gives a reasonable solution...

Why isn't giving me similar solution? Any idea?
 
I agree with CrGreatHouse - your data shouldn't be considered to be normally distributed.

More importantly, if the data in your final post is your entire data set, the sample size is far too small to get a reliable idea of shape of the underlying distribution. A different type of graph (boxplot perhaps) would be a far better choice.
 
What if I use a skew normal distribution... do you think that might fit the data better?
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 24 ·
Replies
24
Views
4K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 25 ·
Replies
25
Views
6K
  • · Replies 6 ·
Replies
6
Views
1K
  • · Replies 1 ·
Replies
1
Views
2K