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


Statistics for the Utterly Confused

This book offers a super-accessible approach to the much-misunderstood subject of statistics.
More information
-->

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

comparing expected vs. actual
Message posted by mike whitehead on December 13, 1999 at 12:00 AM (ET)

I want to generate a statistic that will help me compare two numbers-- an predicted result vs. an actual result. I seem to remember using a % difference calculation in college:
% diff = (|(predicted - actual)| / (predicted)) x 100%.

I have 2 questions:
1) Is this the right formula?, and
2) Is this an appropriate function in the following situation?--

A manager estimates a budget for a project. At the end of the project, the final cost of the project is determined. I want to compare "how good" the prediction was and eventually,
I would like to compare THIS result to other similar projects for this manager, as well as OTHER managers.

I realize that a simple percent would be one alternative [(actual/predicted) x 100%],
but I would like to see what other functions may be appropriate.


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

Re: comparing expected vs. actual
Message posted by JG on December 13, 1999 at 12:00 AM (ET)

I think you should divide by the actual. But this is only a stating point. The 'goodness' of a prediction and meaningful comparisons are complex problems for which there is no clear cut answer.

Statistics



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 咀opyright 1996-2002, Robert Niles.
All rights reserved. Questions? Comments? Read my Privacy Policy, or E-mail me!