rtmPlot {cssTools}

Plots for the ROC Based Threshold Method for Estimating Networks

Description

Visualisation of the ROC based threshold method for estimating networks, implemented by the rtm function.

Usage

rtmPlot(rtmOutput)

Arguments

rtmOutput

Output from the function rtm.

Details

The function rtm uses the density weighted ROC based threshold method (RTM) of Yenigun et. al. (2016) for estimating networks from a random sample of CSS slices. The output from rtm is visualized by the function rtmPlot, which displays the ROC curve, as well as the type 1 and type 2 error counts for each threshold value.

Author(s)

Deniz Yenigun, Gunes Ertan, Michael Siciliano

References

D. Yenigun, G. Ertan, M.D. Siciliano (2016). Omission and commission errors in network cognition and estimation using ROC curve. arXiv:1606.03245 [stat.CO] https://arxiv.org/abs/1606.03245

See Also

rtm

Examples

# Load the highTechManagers data given in cssTools package
data(highTechManagers)

# There are 21 CSS slices in the complete data 
# Suppose we only observed the 10 slices with the following indexes
sampled=c(2,4,5,8,9,10,11,14,18,19)

# Then the observed data is the following
dSampled=highTechManagers[,,sampled]

# Apply the ROC based threshold method to estimate the network
y=rtm(dSampled,sampled)

# Now plot the ROC curve and the error types for various threshold values
rtmPlot(y)