Identifying and removing sources of imprecision in polynomial regression
Identification and removal of imprecision in polynomial regression, originating from random errors (noise) in the independent variable data is discussed. The truncation error-to-noise ratio (TNR) is used to discriminate between imprecision dominated by collinearity, or numerical error propagation, or inflated variance due to noise in the independent variable. It is shown that after the source of the imprecision has been identified, it can often be removed by simple data transformations or using numerical algorithms which are less sensitive to error propagation (such as QR decomposition). In other cases, more precise independent variable data may be required to improve the accuracy and the statistical validity of the correlation.
Year of publication: |
1998
|
---|---|
Authors: | Brauner, Neima ; Shacham, Mordechai |
Published in: |
Mathematics and Computers in Simulation (MATCOM). - Elsevier, ISSN 0378-4754. - Vol. 48.1998, 1, p. 75-91
|
Publisher: |
Elsevier |
Subject: | Regression | Polynomial | Precision | Noise | Collinearity |
Saved in:
Online Resource
Saved in favorites
Similar items by subject
-
Changes in individual and/or aggregate?
Hey, John Denis, (2015)
-
How collinearity affects mixture regression results
Becker, Jan-Michael, (2015)
-
Business performance of airports : non-aviation revenues and their determinants
Fasone, Vincenzo, (2016)
- More ...