Abstract #302105

This is the preliminary program for the 2003 Joint Statistical Meetings in San Francisco, California. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 2-5, 2003); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2003 Program page
JSM 2003 Abstract #302105
Activity Number: 241
Type: Contributed
Date/Time: Tuesday, August 5, 2003 : 10:30 AM to 12:20 PM
Sponsor: Business & Economics Statistics Section
Abstract - #302105
Title: Cost Estimating Relationship Regression Variance Study
Author(s): Donald W. MacKenzie*+
Companies: Wyle Laboratories Inc.
Address: 4461 Den Haag Rd., Warrenton, VA, 20187-2865,
Keywords: cost ; estimating ; relationship ; regression ; variance ; DOF
Abstract:

A previous study by the author revealed that most regression models for space system hardware boxes based on small datasets had relatively low levels of variance. This prompted the current study of relationships between CER variance measures and the number of CER data points. The study is based exclusively on models of the form Y = AX^B, representative of the vast majority of space hardware CERs. Four best-fit methods were evaluated: OLS with both X and Y log-transformed (OLSL), MPE, MUPE, and ZMPE. The model variance (SE, SPE, CV) produced by each approach under the same conditions (a specific set of X-Y data points) was calculated using Monte Carlo sampling techniques for a large number of X-Y data sets and data points per set. Average variance and bias variables were then compared by plotting them against the regression Degrees of Freedom (DOF). Model variance is underestimated by each best-fit method at low DOF, replicating the general behavior of our actual regression models. Therefore, upward adjustment to low-DOF CER variance is recommended for cost risk analyses and model comparisons. Remedial steps for cases with unusually high or low values of B are described.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2003 program

JSM 2003 For information, contact meetings@amstat.org or phone (703) 684-1221. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2003