JCE Online Journal of Chemical Education
 | Subscriptions  | Software Orders  | Support  | Contributors  | Advertisers  | 

JCE Print
   Current Issue
 Previous Issues
 Supplements
 Search JCE Index

JCE Digital Library
   DigiDemos
 QBank
 SymMath
 WebWare

JCE Software
   Latest Releases
 Software & Video
 Downloads
 Support

Only@JCE Online
   JCE Online Store
 JCE HS CLIC
 JCE Discussion Forums
   Biographical Snapshots
 ChemEd Resource Shelf
 Featured Molecules
 Hal's Picks
 Project Chemlab
 Reviewed WWW Sites
 "Web-Ed" Articles

About JCE
   Features
 Publications
 Operations
 Outreach
 Contact Us


  Home > JCE Print > Journal of Chemical Education > Issues > 2005  > January  >
Information • Textbooks • Media • Resources
Understanding Least Squares through Monte Carlo Calculations
Joel Tellinghuisen
Department of Chemistry, Vanderbilt University, Nashville, TN 37235
Cover
January 2005
Vol. 82 No. 1
p. 157
Abstract
The mathematics of linear least squares (LLS) is summarized in a compact and easy-to-remember matrix notation and Monte Carlo computations are used to illustrate its fundamental properties: Gaussian (normal) distributions for parameters, χ2 distributions for variances, and t-distributions for parameters when their variances have to be estimated from the fit. Deviations from these forms are illustrated for cases where the data do not possess random normal error, are biased, or are not weighted properly in the LS fit. The distinction between the a priori and a posteriori variance–covariance matrices is emphasized. The former is appropriate when the error structure of the data is known a priori, as in Monte Carlo calculations; it is exact. The latter assumes ignorance about at least the scale of the data error, which is then estimated from the fit. Importantly, the a posteriori estimates of parameter standard errors are unreliable unless the data are given correct relative weights. The exact equation for error propagation of correlated variables is presented and illustrated. The matter of confidence limits and joint confidence limits on parameters and functions of parameters is discussed and illustrated, with reference to the properties of χ2.
More Information
*  Citation
 Tellinghuisen, Joel. J. Chem. Educ. 2005 82 157.
*  Keywords
 Laboratory Instruction; Numerical Methods; Physical Chemistry; Statistics / Data Analysis
*  History
 Created:
Last Updated:		 November 29, 2004
December 14, 2004
<< Previous Article Table of Contents Next Article >>
  Home > JCE Print > Journal of Chemical Education > Issues > 2005 > January > Page 157


Subscriptions
 Subscription Info
 Order Forms

JCE HS CLIC

Our Secondary School editors work hard to distill all the JCE materials to produce a fraction of particular interest to high school teachers. We call it CLIC.
 JCE HS CLIC

Contributions Welcome
JCE welcomes your submission
 Contributors' Corner

Advertisers
In recent years we have worked hard to better match our advertisers with our readers. When shopping for chemistry education materials, visit our advertisers' WWW sites first.
 Recent Advertisers

Be An Ambassador
Take JCE along on your outreach missions. Copies of the Journal, guest access to JCE Online, our publications catalog, and more are available for your participants.
 Outreach with JCE

Comments to
jceonline@chem.wisc.edu		 Copyright © Division of Chemical Education, Inc., American Chemical Society.
All rights reserved.