Prediction Probe

DistributionProbe

Find the Distribution that Best Fits Your Data

Probabilistic analysis begins by constructing probability distribution models that describe the process input variables and uncertainties. Delivered to the desktop through a familiar Windows-based interface, DistributionProbeTM identifies the best probability distribution model for such variables and uncertainties using available data. DistributionProbeTM  can be used as a standalone tool or in conjunction with UNIPASS®.Version 1.0 key capabilities include:

  • 15 probability distributions
  • 3 goodness-of-fit tests including Kolmogorov-Smirnov (K-S) test, Anderson-Darling test and Cramer Von Mises test to objectively determine the best-fit distribution among the available distributions.
  • 3 methods to estimate distribution parameters including the maximum likelihood method, method of moments, and modified method of moments. 
  •  

     Sample Model

     

  • Subjective tests including probability plots to evaluate the selected distributions.
  • Goodness-of-fit test Critical Value Calculator
  • PDF/CDF Calculator
  • A comprehensive HTML format report containing all pertinent information including, but not limited to, full comparison between various distributions, distribution parameters, PDF and CDF for all selected distributions, probability plots, and more.

 Sample Model

DistributionProbe

DistributionProbe