RobertNiles.com
About Robert
Reporting Help
Finding Data on the Internet
Questions for Candidates
Stats Lessons
Mean
Median
Percent
Per capita
Standard Deviation
Margin of Error
Data Analysis
Sample Sizes
Stats Tests
Help Board
Bookstore


The Cartoon Guide to Statistics

This isn't some dumb-downed whitewash. It's a must-read for a beginner taking his or her first steps toward understanding stats.
More information
-->

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

Lognormal Distribution
Message posted by Paul on December 1, 1999 at 12:00 AM (ET)

I have historical data for variables D and P from which I calculate Y=D/P and the historical average of Y, Yavg. D and P are always positive numbers.
Given the current value of Y, I want to determine the current Y’s statistical deviation from the historical Yavg. Y is lognormal, so I calculate the statistics for the natural logarithm of the historical Y. Then I calculate Z for the current LnY.

If Z = 2, is it valid to say that Y is statistically far from Yavg because LnY is two standard deviations from the average of the historical LnY’s.?

I am concerned about changing the weighting factor on the data by the logarithm transformation. I am interested in Y, not the Ln of Y. Is this a valid concern?


READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Lognormal Distribution
Message posted by SOF on December 2, 1999 at 12:00 AM (ET)

M. J. Moroney in 'facts from figure' on page 139 suggests that a lognormal distribution can be made approximately normal by a logarithmic transformation. After you finish your analysis you would use a antilogarithmic transformation to translate your numbers back to the original data.



Your $5 contribution helps cover part the $500 annual cost of keeping this site online.

Niles Online Sites:RobertNiles.comTheme Park InsiderViolinist.com

RobertNiles.com™, the site, content and services ©Copyright 1996-2002, Robert Niles.
All rights reserved. Questions? Comments? Read my Privacy Policy, or E-mail me!