Finding Bestfit Distribution [image]

  • Context: Undergrad 
  • Thread starter Thread starter hoodrych
  • Start date Start date
  • Tags Tags
    Distribution Image
Click For Summary

Discussion Overview

The discussion revolves around finding the best fit distribution for a dataset, with a focus on using software tools for analysis and graphing. Participants explore different methods and tools, including Excel and CurveExpert, to achieve this goal.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested

Main Points Raised

  • One participant seeks to find the best fit distribution using Excel and expresses a need for graphing capabilities.
  • Another participant suggests using CurveExpert as a tool for finding best fit models and recommends evaluating points from the model for plotting.
  • A participant mentions difficulties with CurveExpert, noting that they have no Y values, only X values, which may be causing errors and resulting in a score of 0.0 for the best fit models.
  • Another participant advises using a cumulative histogram of the data, specifying that the x-axis should represent errors and the y-axis should represent frequency.

Areas of Agreement / Disagreement

Participants express differing experiences and challenges with the tools suggested, indicating that there is no consensus on the best approach or tool for finding the best fit distribution.

Contextual Notes

Some limitations include the lack of Y values for the dataset, which may affect the ability to find a best fit model, and the potential misunderstanding of how to use CurveExpert effectively.

hoodrych
Messages
17
Reaction score
0
I need to find the bestfit distribution for this. I'm using excel, so I need to be able to use xcel to graph it.


[PLAIN]http://img831.imageshack.us/img831/8770/unledcwx.png
 
Last edited by a moderator:
Physics news on Phys.org
I use CurveExpert to find best fit models; give that a try. Then simply evaluate a few points from the model and plot them as separate data.
 
adoado said:
I use CurveExpert to find best fit models; give that a try. Then simply evaluate a few points from the model and plot them as separate data.


Maybe I'm using the program wrong (I don't think I am), but it's not coming up with any results.


I have no Y values, only X's, so that may be the problem? There are lots of errors for almost every test. And when I click the "scan for bestfit" button, the 20 or so different best fit models it tests all come up with a Score of 0.0.
 
As I said in the other thread, you should use a cumulative histogram of the data. The x-axis should be the errors and the y-axis should be the frequency.
 

Similar threads

Undergrad Help me understand skewness in QQ-plots please
  • · Replies 1 ·
Replies
1
Views
2K
Undergrad Relating Moments from one Distribution to the Moments of Another
  • · Replies 7 ·
Replies
7
Views
2K
Undergrad Measurement Error Analysis in Gaussian distribution
  • · Replies 2 ·
Replies
2
Views
2K
Undergrad Finding best fit distribution and its Parameters
  • · Replies 3 ·
Replies
3
Views
4K
Undergrad Adding random numbers: Tolerance analysis
  • · Replies 11 ·
Replies
11
Views
4K
Undergrad Conditional distribution of geometric series
  • · Replies 1 ·
Replies
1
Views
1K
Undergrad Finance: does the volatility of a stock follow a certain distribution?
  • · Replies 24 ·
Replies
24
Views
4K
Undergrad What Is the Probability Atom A Will Emit a Photon Before Atom B?
  • · Replies 4 ·
Replies
4
Views
2K
Undergrad Function to find the probability distribution of a stock price
  • · Replies 5 ·
Replies
5
Views
3K
Undergrad Juicy game probability calculation
  • · Replies 15 ·
Replies
15
Views
2K