Part 1: Deep Dive into Input Distribution Selection - Univariate Data

Logo
Presented by

Dr. Steve Van Drew, Consultant & Trainer

About this talk

This series provides @RISK users with guidance on settings to use in the Fit Distributions to Data dialog box, and insight necessary for narrowing down the often-overwhelming candidate list of probability distributions available in @RISK for modeling inputs. We will examine in-depth: - What settings to use in the multiple tabs and radio buttons/drop-down boxes within the Fit Distributions to Data dialog box - How to know which test result(s) to pay the most attention to - Nuances of the distribution choice that involve both art and science - Subtle differences between the triangular and PERT distributions - How to interview experts, avoid biases associated with subjective probability assessment, and turn interview results into a defensible input distribution Whether a newcomer to @RISK and input distribution selection or an experienced user seeking amplification on these topics, this three-part, deep-dive series was developed with you in mind. Part I: Distribution Fitting of Univariate Data Part I of this series provides in-depth coverage of the @RISK Fit Distributions to Data dialog box settings, the five goodness of fit tests performed when fitting univariate data, and some judgmental aspects of distribution selection that involve both art and science.

Related topics:

More from this channel

Upcoming talks (0)
On-demand talks (228)
Subscribers (50256)
Palisade Company is the world’s leading provider of risk and decision analysis software solutions for science and industry. Our array of software products and custom services enhance the management experience by combining the latest in cutting-edge technology with over 35 years of analytics experience. Palisade’s unified software platform helps clients increase margins, improve performance, expand market share, and maximize operational efficiencies. We have a very simple mission: to minimize risk while maximizing potential.