Classical or 'robust' linear regression? [Klasická nebo robustní lineární regrese?]

Paper


Dohnal, Ludek and Faigl, Paul. 2006. "Classical or 'robust' linear regression? [Klasická nebo robustní lineární regrese?]." 9th Yearly Seminar on Securing Quality of Analytical Results. Komorni Lhotka, Czech Republic 27 - 29 Mar 2006 Czech Republic.
Paper/Presentation Title

Classical or 'robust' linear regression? [Klasická nebo robustní lineární regrese?]

Presentation TypePaper
AuthorsDohnal, Ludek (Author) and Faigl, Paul (Author)
Year2006
Place of PublicationCzech Republic
Web Address (URL) of Paperhttps://ulbld.lf1.cuni.cz/file/213/Klasick%C3%A1%20nebo%20robustn%C3%AD%20line%C3%A1rn%C3%AD%20regrese.pdf
Conference/Event9th Yearly Seminar on Securing Quality of Analytical Results
Event Details
9th Yearly Seminar on Securing Quality of Analytical Results
Event Date
27 to end of 29 Mar 2006
Event Location
Komorni Lhotka, Czech Republic
Abstract

Mathematical processing of the data in Classical Linear Regression Analysis (Least Squares Method) is compared with more 'robust' linear approaches, e.g. the Standardized Principal Components Method and the Regression Method according to Passing & Bablok (Passing-Bablok Regression, 1983).
By 'robust approaches' we understand such computational methods, where there is not possible (or advantageous) to make a distinction between 'independent' and 'dependent' variables. These are fundamentally indistinguishable as a choice and their uncertainties are of a similar order of magnitude. Typically, such is the case of comparison of the data of e.g. two analytical (instrumental) procedures in chemistry. The use of the often applied Least Square Method (LSM) is in such instance inappropriate. In the LSM it is implicitly assumed that the variables have inherently different uncertainty and therefore are not mutually exchangeable.
A comparison between the 3 approaches is presented in a graphical form and open for further discussion.

Keywordsregression analysis, least square method, analytical chemistry, biochemistry, method comparison
ANZSRC Field of Research 2020340199. Analytical chemistry not elsewhere classified
Byline AffiliationsCharles University, Czech Republic
Fibre Composites Design and Development
Permalink -

https://research.usq.edu.au/item/9y5w9/classical-or-robust-linear-regression-klasick-nebo-robustn-line-rn-regrese

Download files

  • 2027
    total views
  • 416
    total downloads
  • 3
    views this month
  • 0
    downloads this month

Export as

Related outputs

Evaluation of the linearity of calibration dependence
Dohnal, Ludek and Faigl, Paul. 2007. "Evaluation of the linearity of calibration dependence." Quality assurance of analytical results (2007). Beskydy, Czech Republic 26 - 28 Mar 2007 Prague, Czech Republic.
Plant based resins for fibre composites
Faigl, Pavel, Rogers, David, Maurin, Romain and Van Erp, Gerard. 2007. "Plant based resins for fibre composites." 4th Annual Composites Australia and Composites CRC Conference (CRC 2007): New Technology Taking Shape. Gold Coast, Australia 19 - 20 Apr 2007 Melbourne, Australia.