Version:	1.3.3.1
Date:	2023-10-04
Title:	Method Comparison Regression
Maintainer:	Sergej Potapov <sergej.potapov@roche.com>
Copyright:	This package includes code from the 'clinfun' library owned by Ventatraman E. Seshan (ktau.r and ktau.f).
Depends:	R (≥ 3.0.0), parallel, robslopes
Imports:	stats, graphics, grDevices, methods
Description:	Regression methods to quantify the relation between two measurement methods are provided by this package. In particular it addresses regression problems with errors in both variables and without repeated measurements. It implements the CLSI recommendations (see J. A. Budd et al. (2018, https://clsi.org/standards/products/method-evaluation/documents/ep09/) for analytical method comparison and bias estimation using patient samples. Furthermore, algorithms for Theil-Sen and equivariant Passing-Bablok estimators are implemented, see F. Dufey (2020, <doi:10.1515/ijb-2019-0157>) and J. Raymaekers and F. Dufey (2022, <arXiv:2202:08060>). A comprehensive overview over the implemented methods and references can be found in the manual pages "mcr-package" and "mcreg".
License:	GPL (≥ 3)
Collate:	"mcrMisc.r" "mcLinReg.r" "mcDeming.r" "mcWDeming.r" "mcPaBaLarge.r" "mcPaBa.r" "mcPBequi.r" "mcCalcCI.r" "mcCalcTstar.r" "mcBootstrap.r" "MCResultMethods.r" "MCResult.r" "MCResultAnalyticalMethods.r" "MCResultAnalytical.r" "MCResultJackknifeMethods.r" "MCResultJackknife.r" "MCResultResamplingMethods.r" "MCResultResampling.r" "MCResultBCaMethods.r" "MCResultBCa.r" "mcrInterface.r" "mcrCompareFit.r" "mcrIncludeLegend.r" "zzz.r" "ktau.r"
RoxygenNote:	7.2.3
LazyData:	true
NeedsCompilation:	yes
Packaged:	2024-09-18 10:24:17 UTC; ripley
Author:	Sergej Potapov [aut, cre], Fabian Model [aut], Andre Schuetzenmeister [aut], Ekaterina Manuilova [aut], Florian Dufey [aut], Jakob Raymaekers [aut], Venkatraman E. Seshan [ctb], Roche [cph, fnd]
Repository:	CRAN
Date/Publication:	2024-09-23 11:15:43 UTC

Method Comparison Regression

Description

Regression methods to quantify the relation between two measurement methods are provided by this package. In particular it addresses regression problems with errors in both variables and without repeated measurements. It implements the CLSI recommendations for analytical method comparison and bias estimation using patient samples.

The main function for performing regression analysis is mcreg. Various functions for summarizing and plotting regression results are provided (see examples in mcreg).

For user site testing (installation verification) please use the test case suite provided with the package. The test case suite can be run by sourcing the 'runalltests.R' script in the 'unitTests' folder. It requires the XML and Runit packages.

Details

Author(s)

Sergej Potapov <sergej.potapov@roche.com>, Fabian Model <fabian.model@roche.com>, Andre Schuetzenmeister <andre.schuetzenmeister@roche.com>, Ekaterina Manuilova <ekaterina.manuilova@roche.com>, Florian Dufey <florian.dufey@roche.com>, Jakob Raymaekers <jakob.raymaekers@kuleuven.com>

References

CLSI EP09 https://clsi.org/

Class "MCResult"

Description

Result of a method comparison.

Objects from the Class

Object is typically created by a call to function mcreg. Object can be directly constructed by calling newMCResult or new("MCResult", data, para, mnames, regmeth, cimeth, error.ratio, alpha, weight).

Slots

data:

Object of class "data.frame" ~~

para:

Object of class "matrix" ~~

mnames:

Object of class "character" ~~

regmeth:

Object of class "character" ~~

cimeth:

Object of class "character" ~~

error.ratio:

Object of class "numeric" ~~

alpha:

Object of class "numeric" ~~

weight:

Object of class "numeric" ~~

Methods

calcBias

signature(.Object = "MCResult"): ...

calcCUSUM

signature(.Object = "MCResult"): ...

calcResponse

signature(.Object = "MCResult"): ...

getCoefficients

signature(.Object = "MCResult"): ...

coef

signature(.Object = "MCResult"): ...

getData

signature(.Object = "MCResult"): ...

getErrorRatio

signature(.Object = "MCResult"): ...

getRegmethod

signature(.Object = "MCResult"): ...

getResiduals

signature(.Object = "MCResult"): ...

getWeights

signature(.Object = "MCResult"): ...

plot

signature(x = "MCResult"): ...

plotBias

signature(x = "MCResult"): ...

plotDifference

signature(.Object = "MCResult"): ...

plotResiduals

signature(.Object = "MCResult"): ...

printSummary

signature(.Object = "MCResult"): ...

summary

signature(.Object = "MCResult"): ...

Author(s)

Ekaterina Manuilova ekaterina.manuilova@roche.com, Andre Schuetzenmeister andre.schuetzenmeister@roche.com, Fabian Model fabian.model@roche.com Sergej Potapov sergej.potapov@roche.com

Examples

showClass("MCResult")

Systematical Bias Between Reference Method and Test Method

Description

Calculate systematical bias between reference and test methods at the decision point Xc as Bias(Xc) = Intercept + (Slope-1) * Xc with corresponding confidence intervals.

Usage

MCResult.calcBias(
  .Object,
  x.levels,
  type = c("absolute", "proportional"),
  percent = TRUE,
  alpha = 0.05,
  ...
)

Arguments

Value

response and corresponding confidence interval for each decision point from x.levels.

See Also

plotBias

Examples

    #library("mcr")
    data(creatinine,package="mcr")
    x <- creatinine$serum.crea
    y <- creatinine$plasma.crea

    # Deming regression fit.
    # The confidence intervals for regression coefficients
    # are calculated with analytical method
    model <- mcreg( x,y,error.ratio = 1,method.reg = "Deming", method.ci = "analytical",

                     mref.name = "serum.crea", mtest.name = "plasma.crea", na.rm=TRUE )
    # Now we calculate the systematical bias
    # between the testmethod and the reference method
    # at the medical decision points 1, 2 and 3 

    calcBias( model, x.levels = c(1,2,3))
    calcBias( model, x.levels = c(1,2,3), type = "proportional")
    calcBias( model, x.levels = c(1,2,3), type = "proportional", percent = FALSE)

Calculate CUSUM Statistics According to Passing & Bablok (1983)

Description

Calculate CUSUM Statistics According to Passing & Bablok (1983)

Usage

MCResult.calcCUSUM(.Object)

Arguments

Value

A list containing the following elements:

References

Passing, H., Bablok, W. (1983) A new biometrical procedure for testing the equality of measurements from two different analytical methods. Application of linear regression procedures for method comparison studies in clinical chemistry, Part I. J Clin Chem Clin Biochem. Nov; 21(11):709–20.

Calculate Response with Confidence Interval.

Description

Calculate Response Intercept + Slope * Refrencemethod with Corresponding Confidence Interval

Usage

MCResult.calcResponse(.Object, x.levels, alpha, ...)

Arguments

Value

response and corresponding confidence interval for each point in vector x.levels.

See Also

calcBias

Examples

    #library("mcr")
    data(creatinine,package="mcr")
    x <- creatinine$serum.crea
    y <- creatinine$plasma.crea
    # Deming regression fit.
    # The confidence intercals for regression coefficients
    # are calculated with analytical method
    model <- mcreg( x,y,error.ratio=1,method.reg="Deming", method.ci="analytical",
                     mref.name = "serum.crea", mtest.name = "plasma.crea", na.rm=TRUE )
    calcResponse(model, x.levels=c(1,2,3))

Get Regression Coefficients

Description

Get Regression Coefficients

Usage

MCResult.getCoefficients(.Object)

Arguments

Value

Regression parameters in matrix form. Rows: Intercept, Slope. Cols: EST, SE, LCI, UCI.

Get Data

Description

Get Data

Usage

MCResult.getData(.Object)

Arguments

Value

Measurement data in matrix format. First column contains reference method (X), second column contains test method (Y).

Get Error Ratio

Description

Get Error Ratio

Usage

MCResult.getErrorRatio(.Object)

Arguments

Value

Error ratio. Only relevant for Deming type regressions.

Get Fitted Values.

Description

This funcion computes fitted values for a 'MCResult'-object. Depending on the regression method and the error ratio, a projection onto the regression line is performed accordingly. For each point (x_i; y_i) i=1,...,n the projected point(x_hat_i; y_hat_i) is computed.

Usage

MCResult.getFitted(.Object)

Arguments

Value

fitted values as data frame.

See Also

plotResiduals getResiduals

Get Regression Method

Description

Get Regression Method

Usage

MCResult.getRegmethod(.Object)

Arguments

Value

Name of the statistical method used for the regression analysis.

Get Regression Residuals

Description

This function returns residuals in x-direction (x-xhat), in y-direction(y-yhat) and optimized residuals. The optimized residuals correspond to distances between data points and the regression line which were optimized for regression coefficients estimation. In case of Passing-Bablok Regression orthogonal residuals will be returned as optimized residuals . The residuals in x-direction are interesting for regression types which assume errors in both variables (deming, weighted deming, Passing-Bablok), particularily for checking of model assumptions.

Usage

MCResult.getResiduals(.Object)

Arguments

Value

residuals as data frame.

See Also

plotResiduals

Get Weights of Data Points

Description

Get Weights of Data Points

Usage

MCResult.getWeights(.Object)

Arguments

Value

Weights of data points.

MCResult Object Initialization

Description

MCResult Object Initialization

Usage

MCResult.initialize(
  .Object,
  data = data.frame(X = NA, Y = NA),
  para = matrix(NA, ncol = 4, nrow = 2),
  mnames = c("unknown", "unknown"),
  regmeth = "unknown",
  cimeth = "unknown",
  error.ratio = 0,
  alpha = 0.05,
  weight = 1
)

Arguments

Value

MCResult object with initialized parameter.

Scatter Plot Method X vs. Method Y

Description

Plot method X (reference) vs. method Y (test) with (optional) line of identity, regression line and confidence bounds for response.

Usage

MCResult.plot(
  x,
  alpha = 0.05,
  xn = 20,
  equal.axis = FALSE,
  xlim = NULL,
  ylim = NULL,
  xaxp = NULL,
  yaxp = NULL,
  x.lab = x@mnames[1],
  y.lab = x@mnames[2],
  add = FALSE,
  draw.points = TRUE,
  points.col = "black",
  points.pch = 1,
  points.cex = 0.8,
  reg = TRUE,
  reg.col = NULL,
  reg.lty = 1,
  reg.lwd = 2,
  identity = TRUE,
  identity.col = NULL,
  identity.lty = 2,
  identity.lwd = 1,
  ci.area = TRUE,
  ci.area.col = NULL,
  ci.border = FALSE,
  ci.border.col = NULL,
  ci.border.lty = 2,
  ci.border.lwd = 1,
  add.legend = TRUE,
  legend.place = c("topleft", "topright", "bottomleft", "bottomright"),
  main = NULL,
  sub = NULL,
  add.cor = TRUE,
  cor.method = c("pearson", "kendall", "spearman"),
  add.grid = TRUE,
  digits = list(coef = 2, cor = 3),
  ...
)

Arguments

Value

No return value, instead a plot is generated

See Also

plotBias, plotResiduals, plotDifference, compareFit,includeLegend

Examples

 library(mcr)
 data(creatinine,package="mcr")
 creatinine <- creatinine[complete.cases(creatinine),]
  x <- creatinine$serum.crea
  y <- creatinine$plasma.crea

  m1 <- mcreg(x,y,method.reg="Deming",  mref.name="serum.crea",
                                        mtest.name="plasma.crea", na.rm=TRUE)
  m2 <- mcreg(x,y,method.reg="WDeming", method.ci="jackknife",
                                        mref.name="serum.crea",
                                        mtest.name="plasma.crea", na.rm=TRUE)

  plot(m1,  xlim=c(0.5,3),ylim=c(0.5,3), add.legend=FALSE,
                           main="Deming vs. weighted Deming regression",
                           points.pch=19,ci.area=TRUE, ci.area.col=grey(0.9),
                           identity=FALSE, add.grid=FALSE, sub="")
  plot(m2, ci.area=FALSE, ci.border=TRUE, ci.border.col="red3",
                           reg.col="red3", add.legend=FALSE,
                           draw.points=FALSE,add=TRUE)

  includeLegend(place="topleft",models=list(m1,m2),
                           colors=c("darkblue","red"), design="1", digits=2)

Plot Estimated Systematical Bias with Confidence Bounds

Description

This function plots the estimated systematical bias ( Intercept + Slope * Refrencemethod ) - Referencemethod with confidence bounds, covering the whole range of reference method X or only part of it.

Usage

MCResult.plotBias(
  x,
  xn = 100,
  alpha = 0.05,
  add = FALSE,
  prop = FALSE,
  xlim = NULL,
  ylim = NULL,
  bias = TRUE,
  bias.lty = 1,
  bias.lwd = 2,
  bias.col = NULL,
  ci.area = TRUE,
  ci.area.col = NULL,
  ci.border = FALSE,
  ci.border.col = NULL,
  ci.border.lwd = 1,
  ci.border.lty = 2,
  zeroline = TRUE,
  zeroline.col = NULL,
  zeroline.lty = 2,
  zeroline.lwd = 1,
  main = NULL,
  sub = NULL,
  add.grid = TRUE,
  xlab = NULL,
  ylab = NULL,
  cut.point = NULL,
  cut.point.col = "red",
  cut.point.lwd = 2,
  cut.point.lty = 1,
  ...
)

Arguments

Value

No return value, instead a plot is generated

See Also

calcBias, plot.mcr, plotResiduals, plotDifference, compareFit

Examples

#library("mcr")
data(creatinine,package="mcr")

creatinine <- creatinine[complete.cases(creatinine),]
x <- creatinine$serum.crea
y <- creatinine$plasma.crea

# Calculation of models
m1 <- mcreg(x,y,method.reg="WDeming", method.ci="jackknife",
                mref.name="serum.crea",mtest.name="plasma.crea", na.rm=TRUE)
m2 <- mcreg(x,y,method.reg="WDeming", method.ci="bootstrap",
                method.bootstrap.ci="BCa",mref.name="serum.crea",
                mtest.name="plasma.crea", na.rm=TRUE)

# Grafical comparison of systematical Bias of two models
plotBias(m1, zeroline=TRUE,zeroline.col="black",zeroline.lty=1,
                ci.area=TRUE,ci.border=FALSE, ci.area.col=grey(0.9),
                main = "Bias between serum and plasma creatinine",
                sub="Comparison of Jackknife and BCa-Bootstrap confidence bounds ")
plotBias(m2, ci.area=FALSE, ci.border=TRUE, ci.border.lwd=2,
                ci.border.col="red",bias=FALSE ,add=TRUE)
includeLegend(place="topleft",models=list(m1,m2), lwd=c(10,2),
                lty=c(2,1),colors=c(grey(0.9),"red"), bias=TRUE,
                design="1", digits=4)

# Drawing of proportional bias
plotBias(m1, ci.area=FALSE, ci.border=TRUE)
plotBias(m1, ci.area=FALSE, ci.border=TRUE, prop=TRUE)
plotBias(m1, ci.area=FALSE, ci.border=TRUE, prop=TRUE, cut.point=0.6)
plotBias(m1, ci.area=FALSE, ci.border=TRUE, prop=TRUE, cut.point=0.6,
             xlim=c(0.4,0.8),cut.point.col="orange", cut.point.lwd=3, main ="")

Bland-Altman Plot

Description

Draw different Bland-Altman plot modifications (see parameter plot.type).

Usage

MCResult.plotDifference(
  .Object,
  xlab = NULL,
  ylab = NULL,
  ref.line = TRUE,
  ref.line.col = "black",
  ref.line.lty = 1,
  ref.line.lwd = 1,
  bias.line.lty = 1,
  bias.line.lwd = 1,
  bias.line.col = "red",
  bias.text.col = NULL,
  bias.text.cex = 0.8,
  loa.line.lty = 2,
  loa.line.lwd = 1,
  loa.line.col = "red",
  loa.text.col = NULL,
  plot.type = 3,
  main = NULL,
  cex = 0.8,
  digits = 2,
  add.grid = TRUE,
  ylim = NULL,
  ...
)

Arguments

Value

No return value, instead a plot is generated

References

Bland, J. M., Altman, D. G. (1986) Statistical methods for assessing agreement between two methods of clinical measurement. Lancet, i: 307–310.

See Also

plot.mcr, plotResiduals, plotDifference, plotBias, compareFit

Examples

    #library("mcr")
    data(creatinine,package="mcr")
    x <- creatinine$serum.crea
    y <- creatinine$plasma.crea

    # Deming regression fit.
    # The confidence intercals for regression coefficients
    # are calculated with analytical method
    model <- mcreg( x,y,error.ratio=1,method.reg="Deming", method.ci="analytical",
                     mref.name = "serum.crea", mtest.name = "plasma.crea", na.rm=TRUE )

    plotDifference( model ) # Default plot.type=3
    plotDifference( model, plot.type=5)
    plotDifference( model, plot.type=7, ref.line.lty=3, ref.line.col="green3" )

Plot Residuals of an MCResult Object

Description

Plot Residuals of an MCResult Object

Usage

MCResult.plotResiduals(
  .Object,
  res.type = c("optimized", "y", "x"),
  xaxis = c("yhat", "both", "xhat"),
  ref.line = TRUE,
  ref.line.col = "red",
  ref.line.lty = 2,
  ref.line.lwd = 1,
  main = NULL,
  xlab = NULL,
  ylab = NULL,
  add.grid = TRUE,
  ...
)

Arguments

Value

No return value, instead a plot is generated

See Also

getResiduals, plot.mcr, plotDifference, plotBias, compareFit

Examples

    data(creatinine,package="mcr")
    x <- creatinine$serum.crea
    y <- creatinine$plasma.crea

    # Deming regression fit.
    # The confidence intercals for regression coefficients
    # are calculated with analytical method
    model <- mcreg( x,y,error.ratio=1,method.reg="WDeming", method.ci="jackknife",
                     mref.name = "serum.crea", mtest.name = "plasma.crea", na.rm=TRUE )
    plotResiduals(model, res.type="optimized", xaxis="both" )
    plotResiduals(model, res.type="y", xaxis="yhat")

Print Summary of a Regression Analysis

Description

Print Summary of a Regression Analysis

Usage

MCResult.printSummary(.Object)

Arguments

Value

No return value

See Also

getCoefficients, getRegmethod

Class "MCResultAnalytical"

Description

Result of a method comparison based on analytical methods for computing confidence intervals.

Objects from the Class

Object is typically created by a call to function mcreg. Object can be directly constructed by calling newMCResultAnalytical or new("MCResultAnalytical", data, xmean, para, mnames, regmeth, cimeth, error.ratio, alpha, weight).

Slots

xmean:

Object of class "numeric" ~~

data:

Object of class "data.frame" ~~

para:

Object of class "matrix" ~~

mnames:

Object of class "character" ~~

regmeth:

Object of class "character" ~~

cimeth:

Object of class "character" ~~

error.ratio:

Object of class "numeric" ~~

alpha:

Object of class "numeric" ~~

weight:

Object of class "numeric" ~~

Extends

Class "MCResult", directly.

Methods

calcResponse

signature(.Object = "MCResultAnalytical"): ...

printSummary

signature(.Object = "MCResultAnalytical"): ...

summary

signature(.Object = "MCResultAnalytical"): ...

Author(s)

Ekaterina Manuilova ekaterina.manuilova@roche.com, Andre Schuetzenmeister andre.schuetzenmeister@roche.com, Fabian Model fabian.model@roche.com, Sergej Potapov sergej.potapov@roche.com

Examples

showClass("MCResultAnalytical")

Caluculate Response

Description

Calculate predicted values for given values of the reference-method.

Usage

MCResultAnalytical.calcResponse(.Object, x.levels, alpha = 0.05)

Arguments

Value

matrix with predicted values with confidence intervals for given values of the reference-method.

Initialize Method for 'MCResultAnalytical' Objects.

Description

Initialize Method for 'MCResultAnalytical' Objects.

Usage

MCResultAnalytical.initialize(
  .Object,
  data = data.frame(X = NA, Y = NA),
  xmean = 0,
  para = matrix(NA, ncol = 4, nrow = 2),
  mnames = c("unknown", "unknown"),
  regmeth = "unknown",
  cimeth = "analytical",
  error.ratio = 0,
  alpha = 0.05,
  weight = 1
)

Arguments

Value

No return value

Print Regression-Analysis Summary for Objects of class 'MCResultAnalytical'.

Description

Function prints a summary of the regression-analysis for objects of class 'MCResultAnalytical'.

Usage

MCResultAnalytical.printSummary(.Object)

Arguments

Value

No return value

Class "MCResultBCa"

Description

Result of a method comparison with BCa-bootstrap based confidence intervals.

Objects from the Class

Object is typically created by a call to function mcreg. Object can be directly constructed by calling newMCResultBCa or new("MCResultBCa", data, para, xmean, mnames, regmeth, cimeth, bootcimeth, alpha, glob.coef, glob.sigma, nsamples, nnested, B0jack, B1jack, B0, B1, MX, rng.seed, rng.kind, sigmaB0, sigmaB1, error.ratio, weight).

Slots

glob.sigma:

Object of class "numeric" ~~

xmean:

Object of class "numeric" ~~

nsamples:

Object of class "numeric" ~~

nnested:

Object of class "numeric" ~~

B0:

Object of class "numeric" ~~

B1:

Object of class "numeric" ~~

sigmaB0:

Object of class "numeric" ~~

sigmaB1:

Object of class "numeric" ~~

MX:

Object of class "numeric" ~~

bootcimeth:

Object of class "character" ~~

rng.seed:

Object of class "numeric" ~~

rng.kind:

Object of class "character" ~~

glob.coef:

Object of class "numeric" ~~

B0jack:

Object of class "numeric" ~~

B1jack:

Object of class "numeric" ~~

data:

Object of class "data.frame" ~~

para:

Object of class "matrix" ~~

mnames:

Object of class "character" ~~

regmeth:

Object of class "character" ~~

cimeth:

Object of class "character" ~~

error.ratio:

Object of class "numeric" ~~

alpha:

Object of class "numeric" ~~

weight:

Object of class "numeric" ~~

Extends

Class "MCResultJackknife", directly. Class "MCResult", by class "MCResultJackknife", distance 2.

Methods

calcResponse

signature(.Object = "MCResultBCa"): ...

printSummary

signature(.Object = "MCResultBCa"): ...

summary

signature(.Object = "MCResultBCa"): ...

Author(s)

Ekaterina Manuilova ekaterina.manuilova@roche.com, Andre Schuetzenmeister andre.schuetzenmeister@roche.com, Fabian Model fabian.model@roche.com, Sergej Potapov sergej.potapov@roche.com

Examples

showClass("MCResultBCa")

Compute Bootstrap-Summary for 'MCResultBCa' Objects.

Description

Function computes the bootstrap summary for objects of class 'MCResultBCa'.

Usage

MCResultBCa.bootstrapSummary(.Object)

Arguments

Value

matrix of bootstrap results

Caluculate Response

Description

Calculate predicted values for given values of the reference-method.

Usage

MCResultBCa.calcResponse(
  .Object,
  x.levels,
  alpha = 0.05,
  bootcimeth = .Object@bootcimeth
)

Arguments

Value

matrix with predicted values with confidence intervals for given values of the reference-method.

Initialize Method for 'MCResultBCa' Objects.

Description

Method initializes newly created objects of class 'MCResultBCa'.

Usage

MCResultBCa.initialize(
  .Object,
  data = data.frame(X = NA, Y = NA),
  para = matrix(NA, ncol = 4, nrow = 2),
  xmean = 0,
  mnames = c("unknown", "unknown"),
  regmeth = "unknown",
  cimeth = "unknown",
  bootcimeth = "unknown",
  alpha = 0.05,
  glob.coef = c(0, 0),
  glob.sigma = c(0, 0),
  nsamples = 0,
  nnested = 0,
  B0jack = 0,
  B1jack = 0,
  B0 = 0,
  B1 = 0,
  MX = 0,
  rng.seed = as.numeric(NA),
  rng.kind = "unknown",
  sigmaB0 = 0,
  sigmaB1 = 0,
  error.ratio = 0,
  weight = 1
)

Arguments

Value

No return value

Plot distriblution of bootstrap coefficients

Description

Plot distriblution of bootstrap coefficients (slope and intercept).

Usage

MCResultBCa.plotBootstrapCoefficients(.Object, breaks = 20, ...)

Arguments

Value

No return value

Plot distriblution of bootstrap pivot T

Description

Plot distriblution of bootstrap pivot T for slope and intercept and compare them with t(n-2) distribution.

Usage

MCResultBCa.plotBootstrapT(.Object, breaks = 20, ...)

Arguments

Value

No return value

Print Regression-Analysis Summary for Objects of class 'MCResultBCa'.

Description

Functions prints a summary of the regression-analysis for objects of class 'MCResultBCa'.

Usage

MCResultBCa.printSummary(.Object)

Arguments

Value

No return value

Class "MCResultJackknife"

Description

Result of a method comparison with Jackknife based confidence intervals.

Objects from the Class

Object is typically created by a call to function mcreg. Object can be directly constructed by calling newMCResultJackknife or new("MCResultJackknife", data, para, mnames, regmeth, cimeth, alpha, glob.coef, B0jack, B1jack, error.ratio, weight).

Slots

glob.coef:

Object of class "numeric" ~~

B0jack:

Object of class "numeric" ~~

B1jack:

Object of class "numeric" ~~

data:

Object of class "data.frame" ~~

para:

Object of class "matrix" ~~

mnames:

Object of class "character" ~~

regmeth:

Object of class "character" ~~

cimeth:

Object of class "character" ~~

error.ratio:

Object of class "numeric" ~~

alpha:

Object of class "numeric" ~~

weight:

Object of class "numeric" ~~

Extends

Class "MCResult", directly.

Methods

calcResponse

signature(.Object = "MCResultJackknife"): ...

getRJIF

signature(.Object = "MCResultJackknife"): ...

plotwithRJIF

signature(.Object = "MCResultJackknife"): ...

printSummary

signature(.Object = "MCResultJackknife"): ...

summary

signature(.Object = "MCResultJackknife"): ...

Author(s)

Ekaterina Manuilova ekaterina.manuilova@roche.com, Andre Schuetzenmeister andre.schuetzenmeister@roche.com, Fabian Model fabian.model@roche.com, Sergej Potapov sergej.potapov@roche.com

Examples

showClass("MCResultJackknife")

Caluculate Response

Description

Calculate predicted values for given values of the reference-method.

Usage

MCResultJackknife.calcResponse(.Object, x.levels, alpha = 0.05)

Arguments

Value

matrix with predicted values with confidence intervals for given values of the reference-method.

Get-Method for Jackknife-Intercept Value.

Description

Extracts the intercept value from objects of class 'MCResultJackknife'.

Usage

MCResultJackknife.getJackknifeIntercept(.Object)

Arguments

Value

(numeric) jackknife-intercept

Get-Method for Jackknife-Slope Value.

Description

Extracts the slope value from objects of class 'MCResultJackknife'.

Usage

MCResultJackknife.getJackknifeSlope(.Object)

Arguments

Value

(numeric) jackknife-slope

Jackknife Statistics

Description

Calculate jackknife mean, bias and standard error.

Usage

MCResultJackknife.getJackknifeStatistics(.Object)

Arguments

Value

table with jackknife mean, bias and standard error for intercept and slope.

Relative Jackknife Influence Function

Description

Calculate the value of relative jackknife function for each observation.

Usage

MCResultJackknife.getRJIF(.Object)

Arguments

Value

a list of the following elements:

References

Efron, B. (1990) Jackknife-After-Bootstrap Standard Errors and Influence Functions. Technical Report , N 134.

Initialize Method for 'MCResultJackknife' Objects.

Description

Method initializes newly created objects of class 'MCResultAnalytical'.

Usage

MCResultJackknife.initialize(
  .Object,
  data = data.frame(X = NA, Y = NA),
  para = matrix(NA, ncol = 4, nrow = 2),
  mnames = c("unknown", "unknown"),
  regmeth = "unknown",
  cimeth = "jackknife",
  alpha = 0.05,
  glob.coef = c(0, 0),
  B0jack = 0,
  B1jack = 0,
  error.ratio = 0,
  weight = 1
)

Arguments

Value

No return value

Plotting the Relative Jackknife Influence Function

Description

The function draws reference method vs. test method as scatter plot. Observations with high influence (relative jackknife influence function is greater than 2) are highlighted as red points.

Usage

MCResultJackknife.plotwithRJIF(.Object)

Arguments

Value

No return value

References

Efron, B. (1990) Jackknife-After-Bootstrap Standard Errors and Influence Functions. Technical Report , N 134.

Examples

    #library("mcr")
    data(creatinine,package="mcr")
    x <- creatinine$serum.crea
    y <- creatinine$plasma.crea
    # Deming regression fit.
    # The confidence intervals for regression coefficients
    # are calculated with jackknife method
    model <- mcreg( x,y,error.ratio=1,method.reg="Deming", method.ci="jackknife",
                     mref.name = "serum.crea", mtest.name = "plasma.crea", na.rm=TRUE )
    plotwithRJIF(model)

Print Regression-Analysis Summary for Objects of class 'MCResultJackknife'.

Description

Functions prints a summary of the regression-analysis for objects of class 'MCResultJackknife'.

Usage

MCResultJackknife.printSummary(.Object)

Arguments

Value

No return value

Class "MCResultResampling"

Description

Result of a method comparison with resampling based confidence intervals.

Objects from the Class

Object is typically created by a call to function mcreg. Object can be directly constructed by calling newMCResultResampling or new("MCResultResampling", data, para, xmean, mnames, regmeth, cimeth, bootcimeth, alpha, glob.coef, rng.seed, rng.kind, glob.sigma, nsamples, nnested, B0, B1, MX, sigmaB0, sigmaB1, error.ratio, weight).

Slots

glob.coef:

Object of class "numeric" ~~

glob.sigma:

Object of class "numeric" ~~

xmean:

Object of class "numeric" ~~

nsamples:

Object of class "numeric" ~~

nnested:

Object of class "numeric" ~~

B0:

Object of class "numeric" ~~

B1:

Object of class "numeric" ~~

sigmaB0:

Object of class "numeric" ~~

sigmaB1:

Object of class "numeric" ~~

MX:

Object of class "numeric" ~~

bootcimeth:

Object of class "character" ~~

rng.seed:

Object of class "numeric" ~~

rng.kind:

Object of class "character" ~~

data:

Object of class "data.frame" ~~

para:

Object of class "matrix" ~~

mnames:

Object of class "character" ~~

regmeth:

Object of class "character" ~~

cimeth:

Object of class "character" ~~

error.ratio:

Object of class "numeric" ~~

alpha:

Object of class "numeric" ~~

weight:

Object of class "numeric" ~~

Extends

Class "MCResult", directly.

Methods

calcResponse

signature(.Object = "MCResultResampling"): ...

printSummary

signature(.Object = "MCResultResampling"): ...

summary

signature(.Object = "MCResultResampling"): ...

Author(s)

Ekaterina Manuilova ekaterina.manuilova@roche.com, Andre Schuetzenmeister andre.schuetzenmeister@roche.com, Fabian Model fabian.model@roche.com, Sergej Potapov sergej.potapov@roche.com

Examples

showClass("MCResultResampling")

Compute Bootstrap-Summary for 'MCResultResampling' Objects.

Description

Function computes the bootstrap summary for objects of class 'MCResultResampling'.

Usage

MCResultResampling.bootstrapSummary(.Object)

Arguments

Value

matrix of bootstrap results

Caluculate Response

Description

Calculate predicted values for given values of the reference-method.

Usage

MCResultResampling.calcResponse(
  .Object,
  x.levels,
  alpha = 0.05,
  bootcimeth = .Object@bootcimeth
)

Arguments

Value

matrix with predicted values with confidence intervals for given values of the reference-method.

Initialize Method for 'MCResultAnalytical' Objects.

Description

Method initializes newly created objects of class 'MCResultAnalytical'.

Usage

MCResultResampling.initialize(
  .Object,
  data = data.frame(X = NA, Y = NA),
  para = matrix(NA, ncol = 4, nrow = 2),
  xmean = 0,
  mnames = c("unknown", "unknown"),
  regmeth = "unknown",
  cimeth = "unknown",
  bootcimeth = "unknown",
  alpha = 0.05,
  glob.coef = c(0, 0),
  rng.seed = as.numeric(NA),
  rng.kind = "unknown",
  glob.sigma = c(0, 0),
  nsamples = 0,
  nnested = 0,
  B0 = 0,
  B1 = 0,
  MX = 0,
  sigmaB0 = 0,
  sigmaB1 = 0,
  error.ratio = 0,
  weight = 1
)

Arguments

Value

No return value

Plot distriblution of bootstrap coefficients

Description

Plot distriblution of bootstrap coefficients (slope and intercept).

Usage

MCResultResampling.plotBootstrapCoefficients(.Object, breaks = 20, ...)

Arguments

Value

No return value

Plot distriblution of bootstrap pivot T

Description

Plot distriblution of bootstrap pivot T for slope and intercept and compare them with t(n-2) distribution.

Usage

MCResultResampling.plotBootstrapT(.Object, breaks = 20, ...)

Arguments

Value

No return value

Print Regression-Analysis Summary for Objects of class 'MCResultResampling'.

Description

Functions prints a summary of the regression-analysis for objects of class 'MCResultResampling'.

Usage

MCResultResampling.printSummary(.Object)

Arguments

Calculate difference between two numeric vectors that gives exactly zero for very small relative differences.

Description

Calculate difference between two numeric vectors that gives exactly zero for very small relative differences.

Usage

calcDiff(X, Y, EPS = 1e-12)

Arguments

Value

difference

Graphical Comparison of Regression Parameters and Associated Confidence Intervals

Description

Graphical comparison of regression parameters (intercept and slope) and their associated 100(1-alpha)% confidence intervals for multiple fitted models of 'MCResult' sub-classes.

Usage

compareFit(...)

Arguments

Value

No return value, instead a plot is generated

Examples

     library("mcr")
     data("creatinine", package="mcr")
     fit.lr <- mcreg(as.matrix(creatinine), method.reg="LinReg", na.rm=TRUE)
     fit.wlr <- mcreg(as.matrix(creatinine), method.reg="WLinReg", na.rm=TRUE)
     compareFit( fit.lr, fit.wlr )

Comparison of blood and serum creatinine measurement

Description

This data set gives the blood and serum preoperative creatinine measurements in 110 heart surgery patients.

Usage

creatinine

Format

A data frame containing 110 observations with serum and plasma creatinin measurements in mg/dL for each sample.

Include Legend

Description

Include legend in regression plot (function plot()) or in bias plot (function plotBias ()) with two or more lines.

Usage

includeLegend(
  models = list(),
  digits = 2,
  design = paste(1:2),
  place = c("topleft", "topright", "bottomleft", "bottomright"),
  colors,
  lty = rep(1, length(models)),
  lwd = rep(2, length(models)),
  box.lty = "blank",
  cex = 0.8,
  bg = "white",
  inset = c(0.01, 0.01),
  bias = FALSE,
  model.names = NULL,
  ...
)

Arguments

Value

Legend in plot.

See Also

plot.mcr, plotBias, plotResiduals, plotDifference, compareFit

Examples

#library("mcr")

 data(creatinine,package="mcr")
 x <- creatinine$serum.crea
 y <- creatinine$plasma.crea

 m1 <- mcreg(x,y,method.reg="Deming", mref.name="serum.crea",
                                        mtest.name="plasma.crea", na.rm=TRUE)
 m2 <- mcreg(x,y,method.reg="WDeming", method.ci="jackknife", 
                                         mref.name="serum.crea",
                                         mtest.name="plasma.crea", na.rm=TRUE)

 plot(m1,  XLIM=c(0.5,3),YLIM=c(0.5,3), Legend=FALSE, 
                          Title="Deming vs. weighted Deming regression", 
                          Points.pch=19,ci.area=TRUE, ci.area.col=grey(0.9),
                          identity=FALSE, Grid=FALSE, Sub="")
 plot(m2, ci.area=FALSE, ci.border=TRUE, ci.border.col="red3", 
                          reg.col="red3", Legend=FALSE,add=TRUE, 
                          Points=FALSE, identity=FALSE, Grid=FALSE)

 includeLegend(place="topleft",models=list(m1,m2), 
                          colors=c("darkblue","red"), design="1", digits=2)

Equivariant Passing-Bablok Regression

Description

This is an implementation of the equivariant Passing-Bablok regression.

Usage

mc.PBequi(
  X,
  Y,
  alpha = 0.05,
  slope.measure = c("radian", "tangent"),
  method.reg = c("PBequi", "TS"),
  extended.output = FALSE,
  calcCI = TRUE,
  methodlarge = TRUE
)

Arguments

Value

a list with elements.

Analytical Confidence Interval

Description

Calculate wald confidence intervals for intercept and slope given point estimates and standard errors.

Usage

mc.analytical.ci(b0, b1, se.b0, se.b1, n, alpha)

Arguments

Value

2x4 matrix of estimates and confidence intervals for intercept and slope.

Resampling estimation of regression parameters and standard errors.

Description

Generate jackknife or (nested-) bootstrap replicates of a statistic applied to data. Only a nonparametric ballanced design is possible. For each sample calculate point estimations and standard errors for regression coefficients.

Usage

mc.bootstrap(
  method.reg = c("LinReg", "WLinReg", "Deming", "WDeming", "PaBa", "PaBaLarge", "TS",
    "PBequi"),
  jackknife = TRUE,
  bootstrap = c("none", "bootstrap", "nestedbootstrap"),
  X,
  Y,
  error.ratio,
  nsamples = 1000,
  nnested = 25,
  iter.max = 30,
  threshold = 1e-08,
  NBins = 1e+06,
  slope.measure = c("radian", "tangent")
)

Arguments

Value

a list consisting of

Author(s)

Ekaterina Manuilova ekaterina.manuilova@roche.com, Fabian Model fabian.model@roche.com, Sergej Potapov sergej.potapov@roche.com

References

Efron, B., Tibshirani, R.J. (1993) An Introduction to the Bootstrap. Chapman and Hall. Carpenter, J., Bithell, J. (2000) Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. Stat Med, 19 (9), 1141–1164.

Student Method for Calculation of Resampling Confidence Intervals

Description

Calculate bootstrap confidence intervals for intercept, slope or bias given a vector of bootstrap point estimates.

Usage

mc.calc.Student(Xboot, xhat, alpha, npoints)

Arguments

Value

a list with elements

References

Carpenter, J., Bithell, J. (2000) Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. Stat Med, 19 (9), 1141–1164.

Bias Corrected and Accelerated Resampling Confidence Interval

Description

Calculate resampling BCa confidence intervals for intercept, slope or bias given a vector of bootstrap and jackknife point estimates.

Usage

mc.calc.bca(Xboot, Xjack, xhat, alpha)

Arguments

Value

a list with elements

References

Carpenter, J., Bithell, J. (2000) Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. Stat Med, 19 (9), 1141–1164.

Quantile Calculation for BCa

Description

We are using the R default (SAS (type=3) seems bugged) quantile calculation instead of the quantile function described in Effron&Tibshirani.

Usage

mc.calc.quant(X, alpha)

Arguments

Value

alpha-quantile of vector X.

Quantile Method for Calculation of Resampling Confidence Intervals

Description

Calculate bootstrap confidence intervals for intercept, slope or bias given the vector of bootstrap point estimates.

Usage

mc.calc.quantile(Xboot, alpha)

Arguments

Value

a list with elements

References

B. Efron and RJ. Tibshirani (1994) An Introduction to the Bootstrap. Chapman & Hall.

Bootstrap-t Method for Calculation of Resampling Confidence Intervals

Description

Calculate resampling confidence intervals for intercept, slope or bias with t-Boot method given a vector of bootstrap point estimates and a vector of bootstrap standard deviations.

Usage

mc.calc.tboot(Xboot, Sboot, xhat, shat, alpha)

Arguments

Value

a list with elements

References

Carpenter, J., Bithell, J. (2000) Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. Stat Med, 19 (9), 1141–1164.

Calculate Matrix of All Pair-wise Slope Angles

Description

This version is implemented in C for computational efficiency.

Usage

mc.calcAngleMat(X, Y, posCor = TRUE)

Arguments

Value

Upper triangular matrix of slopes for all point combinations. Slopes in radian.

Calculate Matrix of All Pair-wise Slope Angles

Description

This is a very slow R version. It should not be called except for debugging purposes.

Usage

mc.calcAngleMat.R(X, Y, posCor = TRUE)

Arguments

Value

Upper triangular matrix of slopes for all point combinations. Slopes in radian.

Jackknife Confidence Interval

Description

Calculate Jackknife confidence intervals for intercept, slope or bias given of vector of jackknife point estimates and global point estimate.

Usage

mc.calcLinnetCI(Xjack, xhat, alpha = 0.05)

Arguments

Value

a list with elements

References

Linnet, K. (1993) Evaluation of Regression Procedures for Methods Comparison Studies. CLIN. CHEM. 39/3, 424–432.

Compute Resampling T-statistic.

Description

Compute Resampling T-statistic. for Calculation of t-Bootstrap Confidence Intervals.

Usage

mc.calcTstar(.Object, x.levels, iter.max = 30, threshold = 1e-06)

Arguments

Value

Tstar numeric vector containing resampling pivot statistic.

References

Carpenter J., Bithell J. Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. Stat Med, 19 (9), 1141-1164 (2000).

Calculate Unweighted Deming Regression and Estimate Standard Errors

Description

Calculate Unweighted Deming Regression and Estimate Standard Errors

Usage

mc.deming(X, Y, error.ratio)

Arguments

Value

a list with elements

References

Linnet K. Evaluation of Regression Procedures for Methods Comparison Studies. CLIN. CHEM. 39/3, 424-432 (1993).

Linnet K. Estimation of the Linear Relationship between the Measurements of two Methods with Proportional Errors. STATISTICS IN MEDICINE, Vol. 9, 1463-1473 (1990).

Calculate ordinary linear Regression and Estimate Standard Errors

Description

Calculate ordinary linear Regression and Estimate Standard Errors

Usage

mc.linreg(X, Y)

Arguments

Value

a list with elements

References

Neter J., Wassermann W., Kunter M. Applied Statistical Models. Richard D. Irwing, INC., 1985.

Returns Results of Calculations in Matrix Form

Description

Returns Results of Calculations in Matrix Form

Usage

mc.make.CIframe(b0, b1, se.b0, se.b1, CI.b0, CI.b1)

Arguments

Value

2x4 matrix of estimates and confidence intervals for intercept and slope.

Passing-Bablok Regression

Description

Passing-Bablok Regression

Usage

mc.paba(
  angM = NULL,
  X,
  Y,
  alpha = 0.05,
  posCor = TRUE,
  calcCI = TRUE,
  slope.measure = c("radian", "tangent")
)

Arguments

Value

Matrix of estimates and confidence intervals for intercept and slope. No standard errors provided by this algorithm.

Passing-Bablok Regression for Large Datasets

Description

This function represents an interface to a fast C-implementation of an adaption of the Passing-Bablok algorithm for large datasets. Instead of building the complete matrix of pair-wise slope values, a pre-defined binning of slope-values is used (Default NBins=1e06). This reduces the required memory dramatically and speeds up the computation.

Usage

mc.paba.LargeData(
  X,
  Y,
  NBins = 1e+06,
  alpha = 0.05,
  posCor = TRUE,
  calcCI = TRUE,
  slope.measure = c("radian", "tangent")
)

Arguments

Value

Matrix of estimates and confidence intervals for intercept and slope. No standard errors provided by this algorithm.

Author(s)

Andre Schuetzenmeister andre.schuetzenmeister@roche.com (partly re-using code of function 'mc.paba')

Examples

 library("mcr")
 data(creatinine,package="mcr")
 
# remove any NAs
crea <- na.omit(creatinine)
    
# call the approximative Passing-Bablok algorithm (Default NBins=1e06)
res1 <- mcreg(x=crea[,1], y=crea[,2], method.reg="PaBaLarge", method.ci="analytical") 
getCoefficients(res1)

# now increase the number of bins and see whether this makes a difference
res2 <- mcreg(x=crea[,1], y=crea[,2], method.reg="PaBaLarge", method.ci="analytical", NBins=1e07) 
getCoefficients(res2)
getCoefficients(res1)-getCoefficients(res2)

Calculate Weighted Deming Regression

Description

Calculate weighted deming regression with iterative algorithm suggested by Linnet. This algorithm is avalaible only for positive values. But even in this case there is no guarantee that the algorithm always converges.

Usage

mc.wdemingConstCV(X, Y, error.ratio, iter.max = 30, threshold = 1e-06)

Arguments

Value

a list with elements

References

Linnet K. Evaluation of Regression Procedures for Methods Comparison Studies. CLIN. CHEM. 39/3, 424-432 (1993).

Linnet K. Estimation of the Linear Relationship between the Measurements of two Methods with Proportional Errors. STATISTICS IN MEDICINE, Vol. 9, 1463-1473 (1990).

Calculate Weighted Ordinary Linear Regression and Estimate Standard Errors

Description

The weights of regression are taken as reverse squared values of the reference method, that's why it is impossible to achieve the calculations for zero values.

Usage

mc.wlinreg(X, Y)

Arguments

Value

a list with elements.

References

Neter J., Wassermann W., Kunter M. Applied Statistical Models. Richard D. Irwing, INC., 1985.

Comparison of Two Measurement Methods Using Regression Analysis

Description

mcreg is used to compare two measurement methods by means of regression analysis. Available methods comprise ordinary and weighted linear regression, Deming and weighted Deming regression and Passing-Bablok regression. Point estimates of regression parameters are computed together with their standard errors and confidence intervals.

Usage

mcreg(
  x,
  y = NULL,
  error.ratio = 1,
  alpha = 0.05,
  mref.name = NULL,
  mtest.name = NULL,
  sample.names = NULL,
  method.reg = c("PaBa", "LinReg", "WLinReg", "Deming", "WDeming", "PaBaLarge", "PBequi",
    "TS"),
  method.ci = c("bootstrap", "jackknife", "analytical", "nestedbootstrap"),
  method.bootstrap.ci = c("quantile", "Student", "BCa", "tBoot"),
  nsamples = 999,
  nnested = 25,
  rng.seed = NULL,
  rng.kind = "Mersenne-Twister",
  iter.max = 30,
  threshold = 1e-06,
  na.rm = FALSE,
  NBins = 1e+06,
  slope.measure = c("radian", "tangent"),
  methodlarge = TRUE
)

Arguments

Details

The regression analysis yields regression coefficients 'Inercept' and 'Slope' of the regression Testmethod = Intercept + Slope * Referencemethod. There are methods for computing the systematical bias between reference and test method at a decision point Xc, Bias(Xc) = Intercept + (Slope-1) * Xc, accompanied by its corresponding standard error and confidence interval. One can use plotting method plotBias for a comprehensive view of the systematical bias.

Weighted regression for heteroscedastic data is available for linear and Deming regression and implemented as a data point weighting with the inverted squared value of the reference method. Therefore calculation of weighted regression (linear and Deming) is available only for positive values (>0). Passing-Bablok regression is only available for non-negative values (>=0).

Confidence intervals for regression parameters and bias estimates are calculated either by using analytical methods or by means of resampling methods ("jackknife", "bootstrap", "nested bootstrap"). An analytical method is available for all types of regression except for weighted Deming. For Passing-Bablok regression the option "analytical" calculates confidence intervals for the regression parameters according to the non-parametric approach given in the original reference.

The "jackknife" (or leave one out resampling) method was suggested by Linnet for calculating confidence intervals of regression parameters of Deming and weighted Deming regression. It is possible to calculate jackknife confidence intervals for all types of regression. Note that we do not recommend this method for Passing-Bablok since it has a tendency of underestimating the variability (jackknife is known to yield incorrect estimates for errors of quantiles).

The bootstrap method requires additionally choosing a value for method.bootstrap.ci. If bootstrap is the method of choice, "BCa", t-bootstrap ("tBoot") and simple "quantile" confidence intervals are recommended (See Efron B. and Tibshirani R.J.(1993),Carpenter J., Bithell J. (2000)). The "nestedbootstrap" method can be very time-consuming but is necessary for calculating t-bootstrap confidence intervals for weighted Deming or Passing-Bablok regression. For these regression methods there are no analytical solutions for computing standard errors, which therefore have to be obtained by nested bootstrapping.

Note that estimating resampling based confidence intervals for Passing–Bablok regressions can take very long time for larger data sets due to the high computational complexity of the algorithm. To mitigate this drawback an adaption of the Passing-Bablok algorithm has been implemented ("PaBaLarge"), which yields approximative results. This approach does not build the complete upper triangular matrix of all 'n*(n-1)/2' slopes. It subdivides the range of slopes into 'NBins' classes, and sorts each slope into one of these bins. The remaining steps are the same as for the exact "PaBa" algorithm, except that these are performed on the binned slopes instead of operating on the matrix of slopes.

Our implementation of exact Passing-Bablok regression ("PaBa") provides two alternative metrics for regression slopes which can result in different regression estimates. As a robust regression method PaBa is essentially invariant to the parameterization of regression slopes, however in the case of an even number of all pairwise slopes the two central slopes are averaged to estimate the final regression slope. In this situation using an angle based metric (slope.measure="radian") will result in a regression estimate that is geometrically centered between the two central slopes, whereas the tangent measure (slope.measure="tangent") proposed in Passing and Bablok (1983) will be geometrically biased towards a higher slope. See below for a pathological example. Note that the difference between the two measures is negligible for data sets with reasonable sample size (N>20) and correlation.

Equivariant Passing-Bablok regression as proposed by Bablok et al. (1988) (see also Dufey 2020) is not bound to slopes near 1 and therefore not only applicable for method comparison but also for method transformation, i.e., when two methods yield results on a different scale. Like ordinary Passing-Bablok regression, the method is robust. This method should be preferred over the older "PaBa" and "PaBalarge" algorithms. Both slope measures "radian" and "tangent" are available as are methods for the determination of confidence intervals -analytical and bootstrap. By default (methodlarge=TRUE), a modified algorithm (Dillencourt et al., 1992) is used which scales quasilinearly and requires little memory. Alternatively (methodlarge=F), a simpler implementation which scales quadratically and is more memory intensive may be called. While point estimates coincide for both implementations, analytic confidence intervals differ slightly. Same holds true for the Theil-Sen estimator, which is a robust alternative to linear regression. Like linear regression, it assumes that x-values are error free.

Value

"MCResult" object containing regression results. The function getCoefficients or printSummary can be used to obtain or print a summary of the results. The function getData allows to see the original data. An S4 object of class "MCResult" containing at least the following slots:

Author(s)

Ekaterina Manuilova ekaterina.manuilova@roche.com, Andre Schuetzenmeister andre.schuetzenmeister@roche.com, Fabian Model fabian.model@roche.com, Sergej Potapov sergej.potapov@roche.com, Florian Dufey florian.dufey@roche.com, Jakob Raymaekers jakob.raymaekers@kuleuven.be

References

Bland, J. M., Altman, D. G. (1986) Statistical methods for assessing agreement between two methods of clinical measurement. Lancet, i: 307–310.

Linnet, K. (1993) Evaluation of Regression Procedures for Methods Comparison Studies. CLIN. CHEM. 39/3, 424–432.

Linnet, K. (1990) Estimation of the Linear Relationship between the Measurements of two Methods with Proportional Errors. Statistics in Medicine, Vol. 9, 1463–1473.

Neter, J., Wassermann, W., Kunter, M. (1985) Applied Statistical Models. Richard D. Irwing, INC.

Looney, S. W. (2010) Statistical Methods for Assessing Biomarkers. Methods in Molecular Biology, vol. 184: Biostatistical Methods. Human Press INC.

Passing, H., Bablok, W. (1983) A new biometrical procedure for testing the equality of measurements from two different analytical methods. Application of linear regression procedures for method comparison studies in clinical chemistry, Part I. J Clin Chem Clin Biochem. Nov; 21(11):709–20.

Bablok, W., Passing, H., Bender, R., & Schneider, B. (1988) A general regression procedure for method transformation. Application of linear regression procedures for method comparison studies in clinical chemistry, Part III. Clinical Chemistry and Laboratory Medicine, 26(11): 783–790.

Dillencourt, M. B., Mount, D. M., & Netanyahu, N. S. (1992) A randomized algorithm for slope selection. International Journal of Computational Geometry & Applications, 2(01): 1–27.

Dufey, F. (2020) Derivation of Passing-Bablok regression from Kendall's tau. The International Journal of Biostatistics, 16(2): 20190157. https://doi.org/10.1515/ijb-2019-0157

Raymaekers, J., Dufey, F. (2022) Equivariant Passing-Bablok regression in quasilinear time. arXiv preprint arXiv:2202.08060. https://doi.org/10.48550/arXiv.2202.08060

Efron, B., Tibshirani, R.J. (1993) An Introduction to the Bootstrap. Chapman and Hall.

Carpenter, J., Bithell, J. (2000) Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. Stat Med, 19 (9), 1141–1164.

CLSI EP9-A2. Method Comparison and Bias Estimation Using Patient Samples; Approved Guideline.

See Also

plotDifference, plot.mcr, getResiduals, plotResiduals, calcResponse, calcBias, plotBias, compareFit

Examples

library("mcr")
data(creatinine,package="mcr")
x <- creatinine$serum.crea
y <- creatinine$plasma.crea
# Deming regression fit.
# The confidence intercals for regression coefficients
# are calculated with analytical method
model1<- mcreg(x,y,error.ratio=1,method.reg="Deming", method.ci="analytical",
               mref.name = "serum.crea", mtest.name = "plasma.crea", na.rm=TRUE)
# Results
printSummary(model1)
getCoefficients(model1)
plot(model1)
# Deming regression fit.
# The confidence intervals for regression coefficients
# are calculated with bootstrap (BCa) method
model2<- mcreg(x,y,error.ratio=1,method.reg="Deming",
               method.ci="bootstrap", method.bootstrap.ci = "BCa",
               mref.name = "serum.crea", mtest.name = "plasma.crea", na.rm=TRUE)
compareFit(model1, model2) 

## Pathological example of Passing-Bablok regression where measure for slope angle matters  
x1 <- 1:10; y1 <- 0.5*x1; x <- c(x1,y1); y <- c(y1,x1) 
m1 <- mcreg(x,y,method.reg="PaBa",method.ci="analytical",slope.measure="radian",
            mref.name="X",mtest.name="Y")
m2 <- mcreg(x,y,method.reg="PaBa",method.ci="analytical",slope.measure="tangent",
            mref.name="X",mtest.name="Y")
plot(m1, add.legend=FALSE,identity=FALSE,
     main="Radian vs. tangent slope measures in Passing-Bablok regression\n(pathological example)",
     ci.area=FALSE,add.cor=FALSE)
plot(m2, ci.area=FALSE,reg.col="darkgreen",reg.lty=2,identity=FALSE,add.legend=FALSE,
     draw.points=FALSE,add=TRUE,add.cor=FALSE)
includeLegend(place="topleft",models=list(m1,m2),model.names=c("PaBa Radian","PaBa Tangent"),
              colors=c("darkblue","darkgreen"),lty=c(1,2),design="1",digits=2)

MCResult Object Constructor with Matrix in Wide Format as Input

Description

MCResult Object Constructor with Matrix in Wide Format as Input

Usage

newMCResult(
  wdata,
  para,
  sample.names = NULL,
  method.names = NULL,
  regmeth = "Unknown",
  cimeth,
  error.ratio,
  alpha = 0.05,
  weight = rep(1, nrow(wdata))
)

Arguments

Value

MCResult object containing regression results.

MCResultAnalytical object constructor with matrix in wide format as input.

Description

MCResultAnalytical object constructor with matrix in wide format as input.

Usage

newMCResultAnalytical(
  wdata,
  para,
  xmean,
  sample.names = NULL,
  method.names = NULL,
  regmeth = "Unknown",
  cimeth = "analytical",
  error.ratio = error.ratio,
  alpha = 0.05,
  weight = rep(1, nrow(wdata))
)

Arguments

Value

MCResultAnalytical object containing regression results.

MCResultBCa object constructor with matrix in wide format as input.

Description

MCResultBCa object constructor with matrix in wide format as input.

Usage

newMCResultBCa(
  wdata,
  para,
  xmean,
  sample.names = NULL,
  method.names = NULL,
  regmeth = "unknown",
  glob.coef,
  glob.sigma,
  cimeth = "unknown",
  bootcimeth = "unknown",
  nsamples,
  nnested,
  rng.seed,
  rng.kind,
  B0jack,
  B1jack,
  B0,
  B1,
  MX,
  sigmaB0,
  sigmaB1,
  error.ratio,
  alpha = 0.05,
  weight = rep(1, nrow(wdata))
)

Arguments

Value

MCResult object containing regression results.

MCResultJackknife Object Constructor with Matrix in Wide Format as Input

Description

MCResultJackknife Object Constructor with Matrix in Wide Format as Input

Usage

newMCResultJackknife(
  wdata,
  para,
  sample.names = NULL,
  method.names = NULL,
  regmeth = "Unknown",
  glob.coef,
  cimeth = "unknown",
  B0jack,
  B1jack,
  error.ratio = error.ratio,
  alpha = 0.05,
  weight = rep(1, nrow(wdata))
)

Arguments

Value

MCResult object containing regression results.

MCResultResampling object constructor with matrix in wide format as input.

Description

MCResultResampling object constructor with matrix in wide format as input.

Usage

newMCResultResampling(
  wdata,
  para,
  xmean,
  sample.names = NULL,
  method.names = NULL,
  regmeth = "unknown",
  glob.coef,
  glob.sigma,
  cimeth = "unknown",
  bootcimeth = "unknown",
  nsamples,
  nnested,
  rng.seed,
  rng.kind,
  B0,
  B1,
  MX,
  sigmaB0,
  sigmaB1,
  error.ratio,
  alpha = 0.05,
  weight = rep(1, nrow(wdata))
)

Arguments

Value

MCResult object containing regression results.