TOC curve R Package 0.0.5

• TOC curve introduction
• Package download and installation
• New features
• Package usage
• Examples
• References

TOC_curve_introduction

The total operating characteristic (TOC) is a statistical method to compare a Boolean variable versus a rank variable. It is a modified model of receiver operating characteristic (ROC) that shows more quantitative information.

TOC can measure the ability of an index variable to diagnose either presence or absence of a characteristic. The diagnosis of presence or absence depends on whether the value of the index is above a threshold. TOC considers multiple possible thresholds. Each threshold generates a two-by-two contingency table, which contains four entries: hits, misses, false alarms, and correct rejections.[1]

The receiver operating characteristic (ROC) also characterizes diagnostic ability. For each threshold, ROC reveals two ratios, hits/(hits + misses) and false alarms/(false alarms + correct rejections), while TOC shows the total information in the contingency table for each threshold.[2]

TOC is applicable to measure diagnostic ability in many fields including but not limited to: land change science, medical imaging, weather forecasting, remote sensing, and materials testing.

Wikipedia Link: https://en.wikipedia.org/wiki/Total_operating_characteristic#cite_note-Si-1

Package_download_and_installation

Install package from GitHub

Install devtoolspackage:

install.packages('devtools')

Import devtools package:

library('devtools')

Install TOCpackage from GitHub:

install_github("Peter-Fisher/TOC")

Import TOC package:

library('TOC')

Install package from local

Download TOC_0.0.5.tar.gz from GitHub

Install package from local download:

install.packages('download_path/TOC_0.0.5.tar.gz', repos = NULL, type="source")

Import TOC package:

library('TOC')

New_features

This package is improved from TOC_0.0-5

• Labels of thresholds are now in the right place.
• Give an option for users to sort the thresholds upward or downward.
• Input parameter thres can modify threshold.
• Allow users to plot multiple curves into the same image and easily compare.

Package_usage

New Input parameters for TOC usage:

Parameters	Options	Descriptions
nthres	numeric (default is NULL)	An optional integer indicating the number of equal-interval thresholds to be evaluated for the TOC curve. See Details below
thres	vector (default is NULL)	An optional numeric vector of thresholds to be evaluated for the TOC curve. See Details below
Sort	‘DECREASE’ ‘INCREASE’ (default is ‘DECREASE’)	Character string indicating whether thresholds sort 'Decrease' or 'Increase'
Thresorder	TRUE FALSE (default is TRUE)	Character string indicating whether variables have clear order
Ordinal	TRUE FALSE (default is FALSE)	Character string indicating whether variables are ordinal
FirstThres	numeric (default is NULL)	Character string indicating whether user want an specified first threshold
LastThres	numeric (default is NULL)	Character string indicating whether user want an specified last threshold
Increment	numeric (default is NULL)	Character string indicating whether user want to specify the increment of thresholds

New Input parameters for TOC curve plot:

Parameters	Options	Descriptions
addAUC	TRUE FALSE (default is TRUE)	An option to show whether to show the AUC on the plot
digitsAUC	numeric (default is 2)	To specified the number of digits when showing the AUC
addCC	TRUE FALSE (default is FALSE)	An option to show whether to show the Correct Corner on the plot
AUClableX	numeric (0-1, default is 0.6)	An option to specified the X Position of AUC labels
AUClableY	numeric (0-1, default is 0.1)	An option to specified the Y Position of AUC labels

Examples

rocd <- ROC(index1, boolean, mask, NAval=0, FirstThres = 10000, LastThres = 50000, Increment = 5000, sort='DECREASE')
plot(rocd, cex=0.8, posL=4)

tocd <- TOC(Prob_Map2, Change_Map2b, MASK4, NAval=0, FirstThres = 10000, LastThres = 50000, Increment = 5000, sort='DECREASE')
plot(tocd, cex=0.8, posL=4, addCC = TRUE, labelThres = TRUE, digitsAUC = 4)

plot(c(tocd1,tocd2), cex=0.8, posL=4,addCC = TRUE, digitsAUC=4)

References

[1] Pontius, Robert Gilmore; Si, Kangping (2014). "The total operating characteristic to measure diagnostic ability for multiple thresholds". International Journal of Geographical Information Science. 28 (3): 570–583. doi:10.1080/13658816.2013.862623.

[2] Pontius Jr, Robert Gilmore; Parmentier, Benoit (2014). "Recommendations for using the Relative Operating Characteristic (ROC)". Landscape Ecology. 29 (3): 367–382. doi:10.1007/s10980-013-9984-8