ValidityTest: Print a formatted table with the validity test for the...

In mechkar: Useful Tools for Scientific Research

Description

This function creates a table showing the most common statistical validity metrics used for testing the validity of a screening test or a predictive model. The way that the output is formated is as follows:

################# Validity Test #######################

			Observed
		+	-
		-------------	---------------	-------------
	Predicted	+ TP	FP	PPV
		a	b	e (e1-e2)
		-------------	---------------	-------------
		- FN	TN	NPV
		c	d	f (f1-f2)
		-------------	---------------	-------------
		Sensitivity	Specificity	Prevalence
		g (g1-g2)	h (h1-h2)	i (i1-i2)
		-------------	---------------	-------------
		Error:	(FP+FN)	/(TP+FP+FN+TN)
		Accuracy:	(TP+TN)	/(TP+FP+FN+TN)
		Precision:	TP	/(TP+FP)
		Recall:	TP	/(TP+FN)
		f1-Score*:	2*(Precision*Recall)	/(Precision+Recall)

* F1-Score: Harmonic mean of precision and recall. ######################################################

Usage

ValidityTest(a,b,c,d,multi=100,caption = "Validity of the Model/Screening")

Arguments

a	the true positive (TP) value
b	the false positive (FP) value
c	the false negative (FN) value
d	the true negative (TN) value
multi	(Optional) The multiplier for the values. The default is 100 for calculating the percentage.
caption	the text to be printed as the title.

Details

The ValidityTest function returns a summary table with the validity metrics most commonly used in epidemiology and in statistical analysis.

Value

A character matrix containing the following statistical metrics:

TP	the true positive value
FP	the false positive value
PPV	the positive predictive value
FN	the false negative value
TN	the true negative value
NPV	the negative predictive value
FN	the false negative value
TN	the true negative value
NPV	the negative predictive value
Sensitivity	the sensitivity of the test
Specificity	the true negative value
Prevalence	the prevalence of the positive cases in the group
Error	the of incorrectly assigned cases
Accuracy	the true negative value
Precision	is the same as the PPV
Recall	Othe name for the sensitivity of the test
F1-Score	the armonic mean of the precision and the recall

All results are given with their confidence intervals.

Author(s)

Tomas Karpati M.D.

Examples

tab1 <- ValidityTest(110,20,80,324)

mechkar documentation built on March 13, 2020, 2:30 a.m.