Title: | Estimation of the Generalized Symmetry Point, an Optimal Cutpoint in Continuous Diagnostic Tests |
---|---|
Description: | Estimation of the cutpoint defined by the Generalized Symmetry point in a binary classification setting based on a continuous diagnostic test or marker. Two methods have been implemented to construct confidence intervals for this optimal cutpoint, one based on the Generalized Pivotal Quantity and the other based on Empirical Likelihood. Numerical and graphical outputs for these two methods are easily obtained. |
Authors: | Mónica López-Ratón [aut, cre], Carmen Cadarso-Suárez [aut], Elisa M. Molanes-López [aut], Emilio Letón [aut] |
Maintainer: | Mónica López-Ratón <[email protected]> |
License: | GPL |
Version: | 1.1.2 |
Built: | 2024-12-24 06:40:29 UTC |
Source: | CRAN |
Continuous biomarkers or diagnostic tests are often used to discriminate between healthy and diseased populations. In clinical practice, it is useful to select an appropriate cutpoint or discrimination value which defines the positive and negative test results.
Several methods for selecting optimal cutpoints in diagnostic tests in the sense of a specific optimality criterion have been proposed in the literature depending on the underlying reason for this choice (see for example, Youden, 1950; Pepe, 2003; Liu, 2012; Rota and Antolini, 2014).
One of the best-known methods is based on the Symmetry point, also known in the literature as the point of equivalence (Greiner et al., 1995; Defreitas et al., 2004; Adlhoch et al., 2011), defined as the point where the sensitivity and specificity measures are equal. Taking into account the costs associated to the false positives and false negatives misclassifications, the Generalized Symmetry point can be defined. This package allows the user to compute the Generalized Symmetry point as the optimal cutpoint for a diagnostic test or continuous marker. The two methods introduced in López-Ratón et al. (2016) for estimating the Generalized Symmetry point and its sensitivity and specificity accuracy measures have been implemented in this package. One of them is based on the Generalized Pivotal Quantity (Weerahandi, 1993; 1995) and the other based on Empirical Likelihood (Thomas and Grunkemeier, 1975). Numerical and graphical outputs for these two methods are easily obtained.
Package: | GsymPoint |
Type: | Package |
Version: | 1.1.2 |
Date: | 2023-10-31 |
License: | GPL |
The GsymPoint package only requires a data-entry file, which must, at minimum, contain the following variables: diagnostic marker; disease status (diseased/healthy); and whether adjustment is to be made for any categorical covariate of interest, a variable that indicates the levels of this covariate. A standard-type data input structure is used, with each row of the database indicating a patient/case and each column referring to a variable.
The most important functions are gsym.point()
, control.gsym.point()
, summary.gsym.point()
and plot.gsym.point()
functions.
The gsym.point()
function computes the Generalized Symmetry point and its corresponding sensitivity and specificity indexes, according to the method selected. The two methods can be chosen simultaneously for estimating the optimal cutpoint.
The control.gsym.point()
function is used to set several parameters that are specific of each method, such as the number of simulations or the constants needed for the Empirical Likelihood method.
The summary.gsym.point()
and plot.gsym.point()
functions produce numerical and graphical outputs, respectively.
Mónica López-Ratón, Carmen Cadarso-Suárez, Elisa M. Molanes-López and Emilio Letón
Maintainer: Mónica López-Ratón <[email protected]>
Adlhoch, C., Kaiser, M., Hoehne, M., Mas Marques, A., Stefas, I., Veas, F. and Ellerbrok, H. (2011). Highly sensitive detection of the group A Rotavirus using Apolipoprotein H-coated ELISA plates compared to quantitative real-time PCR. Virology Journal 8, 63.
Defreitas, G.A., Zimmern, P.E., Lemack, G.E. and Shariat, S.F. (2004). Refining diagnosis of anatomic female bladder outlet obstruction: comparison of pressureflow study parameters in clinically obstructed women with those of normal controls. Urology 64, 675-679.
Greiner, M., Sohr, D. and Göbel, P. (1995). A modified ROC analysis for the selection of cut-off values and the definition of intermediate results of serodiagnostic tests. Journal of Immunological Methods 185, 123-132.
Liu, X. (2012). Classification accuracy and cut point selection. Statistics in Medicine 31(23), 2676-2686.
López-Ratón, M., Cadarso-Suárez, C., Molanes-López, E.M. and Letón, E. (2016). Confidence intervals for the Symmetry point: an optimal cutpoint in continuous diagnostic tests. Pharmaceutical Statistics 15(2), 178-192.
López-Ratón, M., Molanes-López, E.M., Letón, E. and Cadarso-Suárez, C. (2017). GsymPoint: An R Package to Estimate the Generalized Symmetry Point, an Optimal Cut-off Point for Binary Classification in Continuous Diagnostic Tests. The R Journal 9(1), 262-283.
Pepe, M.S. (2003). The statistical evaluation of medical tests for classification and prediction. Oxford University Press, New York.
control.gsym.point is used to set various parameters that control the Generalized Symmetry point computing process.
control.gsym.point(B = 499, c_sampling = 0.25, c_F = 0.25, c_ELq = 0.25, c_R = 0.25, I = 2500)
control.gsym.point(B = 499, c_sampling = 0.25, c_F = 0.25, c_ELq = 0.25, c_R = 0.25, I = 2500)
B |
a numerical value meaningful only in the "EL" method. It specifies the number of simulations in the Empirical Likelihood method. The default value is 499. |
c_sampling |
a numerical value meaningful only in the "EL" method. It specifies the constant needed for resampling in the Empirical Likelihood method. The default value is 0.25. |
c_F |
a numerical value meaningful only in the "EL" method. It specifies the constant needed for estimating the distribution in the Empirical Likelihood method. The default value is 0.25. |
c_ELq |
a numerical value meaningful only in the "EL" method. It specifies the constant needed for estimating the empirical likelihood function in the Empirical Likelihood method. The default value is 0.25. |
c_R |
a numerical value meaningful only in the "EL" method. It specifies the constant needed for estimating the ROC Curve in the Empirical Likelihood method. The default value is 0.25. |
I |
a numerical value meaningful only in the "GPQ" method. It specifies the number of replicates in the Generalized Pivotal Quantity method. The default value is 2500. |
The value yielded by this function is used as the control argument of the gsym.point()
function.
Mónica López-Ratón, Carmen Cadarso-Suárez, Elisa M. Molanes-López and Emilio Letón
library(GsymPoint) data(melanoma) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### # How to set the number of replicates I equal to 2000: gsym.point.GPQ.melanoma<-gsym.point(methods = "GPQ", data = melanoma, marker = "X", status = "group", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(I = 2000), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.melanoma) data(prostate) ########################################################### # Empirical Likelihood Method ("EL") ########################################################### # How to set the number of resamples B equal to 99: gsym.point.GPQ.prostate <- gsym.point (methods = "GPQ", data = prostate, marker = "marker", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(B=99), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.prostate)
library(GsymPoint) data(melanoma) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### # How to set the number of replicates I equal to 2000: gsym.point.GPQ.melanoma<-gsym.point(methods = "GPQ", data = melanoma, marker = "X", status = "group", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(I = 2000), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.melanoma) data(prostate) ########################################################### # Empirical Likelihood Method ("EL") ########################################################### # How to set the number of resamples B equal to 99: gsym.point.GPQ.prostate <- gsym.point (methods = "GPQ", data = prostate, marker = "marker", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(B=99), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.prostate)
The elastase
dataset was obtained from a study carried out in the Cardiology Department at the Galicia General Hospital (Santiago de Compostela, Spain). This study (Amaro et al., 1995) was conducted to assess the clinical usefulness of leukocyte elastase determination in the diagnosis of coronary artery disease (CAD).
data(elastase)
data(elastase)
A data frame with 141 observations on the following 3 variables.
elas
leukocyte elastase. Numerical vector
status
true disease status (absence/presence of coronary artery disease). Numerical vector (0=absence, 1=presence)
gender
patient's gender. Factor with Male
and Female
levels
Amaro, A., Gude, F., González-Juanatey, R., Iglesias, C., Fernández-Vázquez, F., García-Acuña, J. and Gil, M. (1995). Plasma leukocyte elastase concentration in angiographically diagnosed coronary artery disease. European Heart Journal 16, 615–622.
Amaro, A., Gude, F., González-Juanatey, R., Iglesias, C., Fernández-Vázquez, F., García-Acuña, J. and Gil, M. (1995). Plasma leukocyte elastase concentration in angiographically diagnosed coronary artery disease. European Heart Journal 16, 615–622.
data(elastase) summary(elastase)
data(elastase) summary(elastase)
gsym.point is used to construct confidence intervals for the Generalized Symmetry point and its accuracy measures (sensitivity and specificity) for a continuous diagnostic test using two methods: the Generalized Pivotal Quantity (GPQ) method and the Empirical Likelihood (EL) method.
gsym.point (methods, data, marker, status, tag.healthy, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE)
gsym.point (methods, data, marker, status, tag.healthy, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE)
methods |
a character vector selecting the method(s) to be used for estimating the Generalized Symmetry point and its accuracy measures. The possible options are: "GPQ", "EL", "auto", c("GPQ","EL") or c("EL","GPQ"). |
data |
a data frame containing all needed variables: the diagnostic marker, the true disease status and, when it is neccesary, the categorical covariate. |
marker |
a character string with the name of the diagnostic test variable. |
status |
a character string with the name of the variable that distinguishes healthy from diseased individuals. |
tag.healthy |
the value codifying healthy individuals in the |
categorical.cov |
a character string with the name of the categorical covariate according to which the Generalized Symmetry point is to be calculated. The default is NULL (no categorical covariate is considered in the analysis). |
CFN |
a numerical value that specifies the cost of a false negative decision. The default value is 1. |
CFP |
a numerical value that specifies the cost of a false positive decision. The default value is 1. |
control |
output of the |
confidence.level |
a numerical value with the confidence level for the construction of the confidence intervals. The default value is 0.95. |
trace |
a logical value to show information on progress when it is TRUE. The default value is FALSE. |
seed |
a logical value to choose if a seed is fixed for generating the trials in the computation of the confidence intervals in order to reproduce the same simulation process. The default value is FALSE. |
value.seed |
the numerical value for the fixed seed when |
verbose |
a logical value that allows to show extra information on the normality assumption and the Shapiro-Wilk normality p-values. The default value is FALSE. |
The Symmetry point satisfies the equality
, where
and
denote, respectively, the specificity (or true negative fraction) and sensitivity (or true positive fraction). Geometrically, it is the point where the ROC curve and the line
(the perpendicular to the positive diagonal line) intersect, and it can also be seen as the point that maximizes simultaneously both types of correct classifications (Riddle and Stratford, 1999; Gallop et al., 2003) corresponding, therefore, to the probability of correctly classifying any subject, whether it is healthy or diseased (Jiménez-Valverde et al., 2012; 2014).
Taking into account the costs associated to the false positives and false negatives misclassifications, and
, an extension of the Symmetry point called the Generalized Symmetry point,
, can be defined as follows (López-Ratón et al., 2015):
where is the relative loss (cost) of a false positive classification as compared with a false negative classification.
Analogously to the Symmetry point,
is obtained graphically by the intersection point between the ROC curve and the line
.
In this package, the two methods proposed in López-Ratón et al. (2016) for estimating the Generalized Symmetry point and its sensitivity and specificity indexes are available:
"GPQ"
: Method based on the Generalized Pivotal Quantity (Weerahandi, 1993; 1995; Lai et al., 2012). It assumes that the diagnostic test on both groups or a monotone Box-Cox transformation is Normal distributed. So, the Generalized Symmetry point can be estimated from the following equation:
where ,
,
and
denotes the standard Normal cumulative distribution function (cdf), with
and
, i = 0,1, the mean and standard deviation of healthy (
=0) and diseased (
=1) populations, respectively.
To check the assumption of normality, the Shapiro-Wilk test is used with a significance level of 5%.
"EL"
: Method based on the Empirical Likelihood (Thomas and Grunkemeier, 1975). It takes into account that can be seen as two specific quantiles, the
-th quantile of the healthy population and the
-th quantile of the diseased population. Following the same reasoning as in Molanes-López and Letón (2011), and considering that the value of
is known in advance and the Generalized Symmetry point defines an operating point on the ROC curve fulfilling
, the following adjusted empirical log-likelihood ratio function is derived to make inference on
:
where are kernel-type estimates of the cdfs
, of the two populations,
, with
a kernel function and
the smoothing parameter, for
.
"auto"
: the program selects automatically the most appropriate method of the two available, based on the normality assumption. The GPQ is selected under the normality assumption and the EL otherwise.
Returns an object of class "gsym.point" with the following components:
methods |
a character vector with the value of the |
levels.cat |
a character vector indicating the levels of the categorical covariate if the |
call |
the matched call. |
data |
the data frame with the variables used in the call. |
For each of the methods used in the call, a list with the following components is obtained:
"optimal.result" |
a list with the Generalized Symmetry point and its associated sensitivity and specificity accuracy measures with the corresponding confidence intervals. |
"AUC" |
the numerical value of the Area Under the ROC Curve. |
"rho" |
the numerical value of the cost ratio |
"pvalue.healthy" |
the numerical value of the p-value obtained by the Shapiro-Wilk normality test for checking the normality assumption of the marker in the healthy population. |
"pvalue.diseased" |
the numerical value of the p-value obtained by the Shapiro-Wilk normality test for checking the normality assumption of the marker in the diseased population. |
In addition, if the original data are not normally distributed the following components also appears:
"lambda" |
the estimated numerical value of the power used in the Box-Cox transformation. |
"normality.transformed" |
a character string indicating if the transformed marker values by the Box-Cox transformation are normally distributed ("yes") or not ("no"). |
"pvalue.healthy.transformed" |
the numerical value of the p-value obtained by the Shapiro-Wilk normality test for checking the normality assumption of the Box-Cox transformed marker in the healthy population. |
"pvalue.diseased.transformed" |
the numerical value of the p-value obtained by the Shapiro-Wilk normality test for checking the normality assumption of the Box-Cox transformed marker in the diseased population. |
Mónica López-Ratón, Carmen Cadarso-Suárez, Elisa M. Molanes-López and Emilio Letón
Gallop, R.J., Crits-Christoph, P., Muenz, L.R. and Tu, X.M. (2003). Determination and interpretation of the optimal operating point for ROC curves derived through generalized linear models. Understanding Statistics 2, 219-242.
Jiménez-Valverde, A. (2012). Insights into the area under the receiver operating characteristic curve (AUC) as a discrimination measure in species distribution modelling. Global Ecology and Biogeography 21, 498-507.
Jiménez-Valverde, A. (2014). Threshold-dependence as a desirable attribute for discrimination assessment: implications for the evaluation of species distribution models. Biodiversity Conservation 23, 369-385
Lai, C.Y., Tian, L. and Schisterman, E.F. (2012). Exact confidence interval estimation for the Youden index and its corresponding optimal cut-point. Comput. Stat. Data Anal. 56, 1103-1114.
López-Ratón, M., Cadarso-Suárez, C., Molanes-López, E.M. and Letón, E. (2016). Confidence intervals for the Symmetry point: an optimal cutpoint in continuous diagnostic tests. Pharmaceutical Statistics 15(2), 178-192.
López-Ratón, M., Molanes-López, E.M., Letón, E. and Cadarso-Suárez, C. (2017). GsymPoint: An R Package to Estimate the Generalized Symmetry Point, an Optimal Cut-off Point for Binary Classification in Continuous Diagnostic Tests. The R Journal 9(1), 262-283.
Metz, C.E. (1978). Basic Principles of ROC Analysis. Seminars in Nuclear Medicine 8, 183-298.
Molanes-López, E.M. and Letón, E. (2011). Inference of the Youden index and associated threshold using empirical likelihood for quantiles. Statistics in Medicine 30, 2467-2480.
Molanes-López, E.M., Van Keilegom, I. and Veraverbeke, N. (2009). Empirical likelihood for non-smooth criterion functions. Scandinavian Journal of Statistics 36, 413-432.
Remaley, A.T., Sampson, M.L., DeLeo, J.M., Remaley, N.A., Farsi, B.D. and Zweig, M.H. (1999). Prevalence-value-accuracy plots: a new method for comparing diagnostic tests based on misclassification costs. Clinical Chemistry 45, 934-941.
Riddle, D.L. and Stratford, P.W. (1999). Interpreting validity indexes for diagnostic tests: An illustration using the Berg Balance Test. Physical Therapy 79, 939-948.
Rutter, C.M. and Miglioretti, D.L. (2003). Estimating the accuracy of psychological scales using longitudinal data. Biostatistics 4, 97-107.
Thomas, D.R. and Grunkemeier, G.L. (1975). Confidence interval estimation of survival probabilities for censored data. Journal of the American Statistical Association 70, 865-871.
Wand, M.P. and Jones, M.C. (1995). Kernel smoothing. Chapman and Hall, London.
Weerahandi, S. (1993). Generalized confidence intervals. Journal of the American Statistical Association 88, 899-905.
Weerahandi, S. (1995). Exact statistical methods for data analysis. Springer-Verlag, New York.
Zhou, W. and Jing, B.Y. (2003). Adjusted empirical likelihood method for quantiles. Annals of the Institute of Statistical Mathematics 55, 689-703.
control.gsym.point
, summary.gsym.point
library(GsymPoint) data(melanoma) ########################################################### # marker: X # status: group ########################################################### ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): # Original data normally distributed ########################################################### gsym.point.GPQ.melanoma<-gsym.point(methods = "GPQ", data = melanoma, marker = "X", status = "group", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(),confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.melanoma) plot(gsym.point.GPQ.melanoma) data(prostate) ########################################################### # marker: marker # status: status ########################################################### ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): # Box-Cox transformed data normally distributed ########################################################### gsym.point.GPQ.prostate <- gsym.point (methods = "GPQ", data = prostate, marker = "marker", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.prostate) plot(gsym.point.GPQ.prostate) data(elastase) ########################################################### # marker: elas # status: status ########################################################### ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): # Original data not normally distributed # Box-Cox transformed data not normally distributed ########################################################### gsym.point.GPQ.elastase <- gsym.point(methods = "GPQ", data = elastase, marker = "elas", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.elastase) plot(gsym.point.GPQ.elastase)
library(GsymPoint) data(melanoma) ########################################################### # marker: X # status: group ########################################################### ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): # Original data normally distributed ########################################################### gsym.point.GPQ.melanoma<-gsym.point(methods = "GPQ", data = melanoma, marker = "X", status = "group", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(),confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.melanoma) plot(gsym.point.GPQ.melanoma) data(prostate) ########################################################### # marker: marker # status: status ########################################################### ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): # Box-Cox transformed data normally distributed ########################################################### gsym.point.GPQ.prostate <- gsym.point (methods = "GPQ", data = prostate, marker = "marker", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.prostate) plot(gsym.point.GPQ.prostate) data(elastase) ########################################################### # marker: elas # status: status ########################################################### ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): # Original data not normally distributed # Box-Cox transformed data not normally distributed ########################################################### gsym.point.GPQ.elastase <- gsym.point(methods = "GPQ", data = elastase, marker = "elas", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.elastase) plot(gsym.point.GPQ.elastase)
The melanoma
dataset was obtained from Venkatram and Begg (1996). This study was conducted to assess the clinical usefulness of a clinical scoring scheme without dermoscope (CSS) used by dermatologists in order to determine if a suspicious pigmented lesion on the skin is a melanoma.
data(melanoma)
data(melanoma)
A data frame with 72 observations on the following 2 variables.
X
clinical scoring scheme without dermoscope. Numerical vector.
group
true disease status (absence/presence of melanoma). Numerical vector (0=absence, 1=presence)
Venkatraman, ES., Begg, CB. (1996). A distribution-free procedure for comparing receiver operating characteristic curves from a paired experiment. Biometrika 83, 835–848.
Venkatraman, ES., Begg, CB. (1996). A distribution-free procedure for comparing receiver operating characteristic curves from a paired experiment. Biometrika 83, 835–848.
data(melanoma) summary(melanoma)
data(melanoma) summary(melanoma)
On the basis of a gsym.point
object, it is used to plot the Receiver Operating Characteristic (ROC) curve, the line and the optimal ROC coordinates associated to the Generalized Symmetry point.
## S3 method for class 'optimal.cutpoints' ## S3 method for class 'gsym.point' plot(x, xlab, ylab, main, ...)
## S3 method for class 'optimal.cutpoints' ## S3 method for class 'gsym.point' plot(x, xlab, ylab, main, ...)
x |
an object of class |
xlab |
the x axis label of the plot. By default this label is set to "False Positive Rate" |
ylab |
the y axis label of the plot. By default this label is set to "True Positive Rate" |
main |
the title of the plot. By default this title is set to "Empirical ROC Curve and line |
... |
further arguments passed to or from other methods |
Mónica López-Ratón, Carmen Cadarso-Suárez, Elisa M. Molanes-López and Emilio Letón
gsym.point
, control.gsym.point
library(GsymPoint) data(melanoma) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.melanoma<-gsym.point(methods = "GPQ", data = melanoma, marker = "X", status = "group", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(),confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) plot(gsym.point.GPQ.melanoma) data(prostate) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.prostate <- gsym.point (methods = "GPQ", data = prostate, marker = "marker", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) plot(gsym.point.GPQ.prostate) data(elastase) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.elastase <- gsym.point(methods = "GPQ", data = elastase, marker = "elas", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) plot(gsym.point.GPQ.elastase)
library(GsymPoint) data(melanoma) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.melanoma<-gsym.point(methods = "GPQ", data = melanoma, marker = "X", status = "group", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(),confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) plot(gsym.point.GPQ.melanoma) data(prostate) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.prostate <- gsym.point (methods = "GPQ", data = prostate, marker = "marker", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) plot(gsym.point.GPQ.prostate) data(elastase) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.elastase <- gsym.point(methods = "GPQ", data = elastase, marker = "elas", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) plot(gsym.point.GPQ.elastase)
Default print method for objects fitted with gsym.point()
function. A short summary is printed with: the call to the gsym.point()
function for each categorical covariate level
(if the categorical.cov
argument of the gsym.point()
function is not NULL).
## S3 method for class 'gsym.point' print(x, digits = max(3L, getOption("digits") - 3L), ...)
## S3 method for class 'gsym.point' print(x, digits = max(3L, getOption("digits") - 3L), ...)
x |
an object of class |
digits |
controls number of digits printed in the output. |
... |
further arguments passed to or from other methods. |
Mónica López-Ratón, Carmen Cadarso-Suárez, Elisa M. Molanes-López and Emilio Letón
gsym.point
, summary.gsym.point
library(GsymPoint) data(elastase) ########################################################### # Empirical Likelihood Method ("GPQ"): ########################################################### gsym.point.GPQ.elastase<-gsym.point(methods = "GPQ", data = elastase, marker = "elas", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) gsym.point.GPQ.elastase print(gsym.point.GPQ.elastase)
library(GsymPoint) data(elastase) ########################################################### # Empirical Likelihood Method ("GPQ"): ########################################################### gsym.point.GPQ.elastase<-gsym.point(methods = "GPQ", data = elastase, marker = "elas", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) gsym.point.GPQ.elastase print(gsym.point.GPQ.elastase)
The prostate
dataset was obtained from a study carried in Le (2006). This study was conducted to evaluate the clinical usefulness of the level of acid phosphatase in blood serum (APBS) x 100 to predict whether
nodal involvement is present.
data(prostate)
data(prostate)
A data frame with 53 observations on the following 2 variables.
marker
level of acid phosphatase in blood serum. Numeric vector
status
true disease status (absence/presence of nodal involvement). Numerical vector (0=absence, 1=presence)
Le, C.T. (2006). A solution for the most basic optimization problem associated with an ROC curve. Statistical Methods in Medical Research 15, 571–584.
Le, C.T. (2006). A solution for the most basic optimization problem associated with an ROC curve. Statistical Methods in Medical Research 15, 571–584.
data(prostate) summary(prostate)
data(prostate) summary(prostate)
Produces a summary of a gsym.point
object. The following is printed: the matched call to the gsym.point()
main function; the area under the ROC curve (AUC) estimate; the Generalized Symmetry point obtained with the method(s) selected and the point estimates of the associated sensitivity and specificity indexes with their corresponding confidence intervals. All this information will be shown for each categorical covariate level (if the categorical.cov
argument in the gsym.point()
function is not NULL).
## S3 method for class 'gsym.point' summary(object, ...)
## S3 method for class 'gsym.point' summary(object, ...)
object |
an object of class |
... |
further arguments passed to or from other methods. None are used in this method. |
The summary.gsym.point
function produces a list of summary information for a fitted gsym.point
object. The result depends on the two arguments, namely, methods
, and categorical.cov
of the gsym.point()
function used in the Generalized Symmetry point computing process.
Returns an object of class "summary.gsym.point" with the same components as the gsym.point
function (see gsym.point
)
Mónica López-Ratón, Carmen Cadarso-Suárez, Elisa M. Molanes-López and Emilio Letón
library(GsymPoint) data(melanoma) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.melanoma<-gsym.point(methods = "GPQ", data = melanoma, marker = "X", status = "group", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(),confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.melanoma) data(prostate) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.prostate <- gsym.point (methods = "GPQ", data = prostate, marker = "marker", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.prostate) data(elastase) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.elastase <- gsym.point(methods = "GPQ", data = elastase, marker = "elas", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.elastase)
library(GsymPoint) data(melanoma) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.melanoma<-gsym.point(methods = "GPQ", data = melanoma, marker = "X", status = "group", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(),confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.melanoma) data(prostate) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.prostate <- gsym.point (methods = "GPQ", data = prostate, marker = "marker", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.prostate) data(elastase) ########################################################### # Generalized Pivotal Quantity Method ("GPQ"): ########################################################### gsym.point.GPQ.elastase <- gsym.point(methods = "GPQ", data = elastase, marker = "elas", status = "status", tag.healthy = 0, categorical.cov = NULL, CFN = 1, CFP = 1, control = control.gsym.point(), confidence.level = 0.95, trace = FALSE, seed = FALSE, value.seed = 3, verbose = FALSE) summary(gsym.point.GPQ.elastase)