Search
Browse by Subject
Contact Information

Northern Research Station
11 Campus Blvd., Suite 200
Newtown Square, PA 19073
(610) 557-4017
(610) 557-4132 TTY/TDD

You are here: NRS Home / Publications & Data / Publication Details
Publication Details

Title: The Sensitivity of Derived Estimates to the Measurement Quality Objectives for Independent Variables

Author: Roesch, Francis A.

Year: 2005

Publication: In: McRoberts, Ronald E.; Reams, Gregory A.; Van Deusen, Paul C.; McWilliams, William H.; Cieszewski, Chris J., eds. Proceedings of the fourth annual forest inventory and analysis symposium; Gen. Tech. Rep. NC-252. St. Paul, MN: U.S. Department of Agriculture, Forest Service, North Central Research Station. 173-180

Abstract: The effect of varying the allowed measurement error for individual tree variables upon county estimates of gross cubic-foot volume was examined. Measurement Quality Objectives (MQOs) for three forest tree variables (biological identity, diameter, and height) used in individual tree gross cubic-foot volume equations were varied from the current USDA Forest Service Forest Inventory and Analysis specifications in a simulation under alternative error models. Assuming unbiased errors may lead to a different control strategy than assuming unbiased errors. Strengthening the MQO for diameter was shown to help reduce the overall variance of volume estimates if diameter errors are slightly biased. Height errors responded favorably to increased control under both the biased and unbiased models. County volume estimates are somewhat robust to the MQOs for biological identity. However, increased control of biological identity did play a more important role when the underlying distributions for diameter and height were assumed to be biased than when these errors were assumed to be unbiased.

Last Modified: 8/11/2006


Publication Toolbox

This document is in PDF format. You can obtain a free PDF reader from Adobe.