EasyFit – Easily Fit Distributions to Your Data!
EasyFit Help HomeLearn More About EasyFit

Manual Distribution Fitting

While the automated distribution fitting is very useful for most applications, sometimes you might need to fit the predetermined models to your data. EasyFit aids in this task by providing a way to manually specify distribution parameters and see how the output results (graphs and tables) change depending on the parameter values.

To take advantage of this feature, you first need to make sure the distribution parameters window is visible. When the fitting results window is active, select the Graphs page, and enable the Tools|Options|Show Parameters option.

When you select a particular distribution, it is displayed in a separate window allowing to manually specify parameter values. Click the Apply Changes icon or press Enter for the changes to take effect. As a result, EasyFit will update the associated graphs and tables.

To fit an additional distribution to your data, select the Tools|Add Distribution menu option. EasyFit will automatically activate and estimate parameters of the newly added distribution.

To remove one or more distributions from the list, select Tools|Remove Distribution.

The distribution parameters window also enables you to change some fitting options (specifically, domain bounds) and rerun the fit. If the Fix/Unfix Parameter button is pressed, the corresponding parameter will not be estimated – it will be assigned the value you specify. The rest of the parameters will be estimated. To rerun the fit, click the Estimate Parameters icon.

To learn more on how the fixed/unfixed parameters affect the distribution fitting process, see the Domain Bounds section.

Additional features available in this window:

  • Select Distribution allows to select another probability distribution which will replace the current one;
  • Explore Using StatAssist displays the current distribution in the StatAssist window;
  • Show Help shows help on the selected probability distribution.
Copyright © MathWave Technologies