You are here:Home>Lognormal Conversion From Stat::Fit

Lognormal Conversion From Stat::Fit

When I generate a lognormal distribution in Stat::Fit it has 3 parameters. But when I copy and paste it to ProcessModel it only has 2. What do the parameters mean?


For the Lognormal distribution, Stat::Fit uses the parameters [min, mu, sigma] where min is the offset from 0, and mu, sigma are the mean and standard deviation of the included normal distribution. ProcessModel uses mean, sd which are the mean and standard deviation of the data after the offset is subtracted. The conversion is:

mean = exp(mu + (sigma^2)/2)
sd = exp(2*mu + sigma^2)*(exp(sigma^2) – 1)

You can find this and more in ”Simulation Modeling & Analysis”, Law & Kelton, 3 rd ed., 2000, p307.

Please note that Stat::Fit will do the conversion. Just open the Distribution Viewer for Lognormal, enter one set of parameters and the other set will be there as well. Also, the Distribution Viewer can be used to export to ProcessModel.

If you need further help, please contact support.

By |Comments Off on Lognormal Conversion From Stat::Fit

About the Author:

Scott Baird has been president of ProcessModel for more than 15 years. His focus has been to teach others how to improve processes dramatically. He has been successful in transferring these skills to over 200 companies, including ESPN, NASA, GE, Nationwide, Cendant, SSA and many more. Specialties: Group facilitation for process improvement, process design and simulation, simulation modeling, business management and training others to see opportunities. Scott loves to teach process improvement and has often been heard to say, “Of all the things I do, training others to improve processes is my favorite.” Scott is a father of four and a grandfather of eight. He is an avid woodworker, designing and creating presentation boxes. In his spare time, he volunteers in a college preparation program.