A basic computer program for evaluating the effect of sample distribution on the least-squares regression slope estimate
✍ Scribed by Curtis A. Parvin
- Publisher
- Elsevier Science
- Year
- 1985
- Tongue
- English
- Weight
- 439 KB
- Volume
- 18
- Category
- Article
- ISSN
- 0010-4809
No coin nor oath required. For personal study only.
✦ Synopsis
Least-squares regression analysis is widely used in analytical method comparison studies even though model assumptions are typically violated. The advantages favoring the use of the least-squares technique, when applicable, are that its theoretical characteristics are thoroughly developed and the calculations are straightforward and universally known. In order to aid an investigator in determining whether or not the least-squares technique can be acceptably applied in a given method comparison situation, a BASIC computer program is developed that evaluates the bias of the least-squares slope estimate as a function of the precision characteristics of the analytical methods and the proposed sampling distribution for the method comparison experiment.