Weighted reduced major axis method for regression model
Paper
Paper/Presentation Title | Weighted reduced major axis method for regression model |
---|---|
Presentation Type | Paper |
Authors | Saqr, Anwar (Author) and Khan, Shahjahan (Author) |
Editors | Ahmad, Munir |
Journal or Proceedings Title | Proceedings of the 12th Islamic Countries Conference on Statistical Sciences |
Journal Citation | 23, pp. 61-70 |
Number of Pages | 10 |
Year | 2012 |
Place of Publication | Lahore, Pakistan |
ISBN | 9789698858117 |
Web Address (URL) of Paper | https://old.isoss.net/downloads/Prociccs12.pdf |
Web Address (URL) of Conference Proceedings | https://isoss.net/organization-2/ |
Conference/Event | 12th Islamic Countries Conference on Statistical Sciences |
Event Details | 12th Islamic Countries Conference on Statistical Sciences Parent Islamic Countries Conference on Statistical Sciences (ICCS) Delivery In person Event Date 19 to end of 22 Dec 2012 Event Location Doha, Qatar |
Abstract | The reduced major axis (RMA) method is widely used in many disciplines as a solution to errors in variables model, although it lacks efficiency. This paper provides an alternative view on RMA estimator. Moreover, it introduces a new estimator to fit regression line when both variables are subject to measurement errors. The proposed weighted reduced major axis (WR) estimator is derived based on the mathematical relationship between the vertical and orthogonal distances of the observed points and the regression line. Compared to the RMA and OLS-bisector estimators the proposed WR estimator is less sensitive to the variation of the ratio of error variances. The simulation results show that the WR estimator is more consistent and efficient than the RMA and OLS-bisector estimators. |
Keywords | linear regression models; measurement error models; reflection of points; ratio of error variances; OLS-bisector |
ANZSRC Field of Research 2020 | 490502. Biostatistics |
490411. Real and complex functions (incl. several variables) | |
490103. Calculus of variations, mathematical aspects of systems theory and control theory | |
Byline Affiliations | Department of Mathematics and Computing |
Institution of Origin | University of Southern Queensland |
https://research.usq.edu.au/item/q21x0/weighted-reduced-major-axis-method-for-regression-model
Download files
Published Version
Saqr_Khan_ICCS2012_PV.pdf | ||
License: CC BY-NC 4.0 | ||
File access level: Anyone |
Other Documentation
ICCS2012.pdf | ||
File access level: Anyone |
2336
total views411
total downloads3
views this month0
downloads this month