Tolerance Interval and EIV Regression - Method Comparison Studies
Back transforms the results if a logarithmic transformation is used
Tolerance Intervals and Errors-in-Variables Regressions in Method Comp...
Fit a Bivariate Least Square regression (BLS): estimates table
Bivariate Least Square regression (BLS)
Bivariate Least Square regression (BLS)
Fit a Correlated Bivariate Least Square regression (CBLS): estimates t...
Correlated Bivariate Least Square regression (CBLS)
Descriptive statistics in method comparison studies
Degrees of freedom by Welch-Satterthwaite
Deming Regression
Confidence Intervals by CBLS with all potential solutions
Confidence Intervals from OLSv to OLSh by DR and BLS
Plot all the CBLS potential solutions
Plot all the DR and BLS potential solutions
Measurement error variances ratio
Tolerance intervals in a (M,D) plot
Display the CBLS regression, or univariate tolerance intervals in a (M...
Horizontal Ordinary Least Square regression
Vertical Ordinary Least Square regression
Raw plot for descriptive statistics
Display the BLS regression in a (X,Y) plot
Assess the agreement in method comparison studies by tolerance intervals and errors-in-variables (EIV) regressions. The Ordinary Least Square regressions (OLSv and OLSh), the Deming Regression (DR), and the (Correlated)-Bivariate Least Square regressions (BLS and CBLS) can be used with unreplicated or replicated data. The BLS() and CBLS() are the two main functions to estimate a regression line, while XY.plot() and MD.plot() are the two main graphical functions to display, respectively an (X,Y) plot or (M,D) plot with the BLS or CBLS results. Four hyperbolic statistical intervals are provided: the Confidence Interval (CI), the Confidence Bands (CB), the Prediction Interval and the Generalized prediction Interval. Assuming no proportional bias, the (M,D) plot (Band-Altman plot) may be simplified by calculating univariate tolerance intervals (beta-expectation (type I) or beta-gamma content (type II)). Major updates from last version 1.0.0 are: title shortened, include the new functions BLS.fit() and CBLS.fit() as shortcut of the, respectively, functions BLS() and CBLS(). References: B.G. Francq, B. Govaerts (2016) <doi:10.1002/sim.6872>, B.G. Francq, B. Govaerts (2014) <doi:10.1016/j.chemolab.2014.03.006>, B.G. Francq, B. Govaerts (2014) <http://publications-sfds.fr/index.php/J-SFdS/article/view/262>, B.G. Francq (2013), PhD Thesis, UCLouvain, Errors-in-variables regressions to assess equivalence in method comparison studies, <https://dial.uclouvain.be/pr/boreal/object/boreal%3A135862/datastream/PDF_01/view>.