table.stats {verification}

Verification statistics for a 2 by 2 Contingency Table

Description

Provides a variety of statistics for a data summarized in a 2 by 2 contingency table.

Usage

       	table.stats(obs, pred, fudge = 0.01, silent = FALSE)

Arguments

obs

Either a vector of contingency table counts, a vector of binary observations, or a 2 by 2 matrix in the form of a contingency table. (See note below.)

pred

Either null or a vector of binary forecasts.

fudge

A numeric fudge factor to be added to each cell of the contingency table in order to avoid division by zero.

silent

Should warning statements be surpressed.

Value

tab.out

Contingency table

TS

Threat score a.k.a. Critical success index (CSI)

TS.se

Standard Error for TS

POD

Hit Rate aka probability of detection

POD.se

Standard Error for POD

M

Miss rate

F

False Alarm RATE

F.se

Standard Error for F

FAR

False Alarm RATIO

FAR.se

Standard Error for FAR

HSS

Heidke Skill Score

HSS.se

Standard Error for HSS

PSS

Peirce Skill Score

PSS.se

Standard Error for PSS

KSS

Kuiper's Skill Score

PC

Percent correct - events along the diagonal.

PC.se

Standard Error for PC

BIAS

Bias

ETS

Equitable Threat Score

ETS.se

Standard Error for ETS

theta

Odds Ratio

log.theta

Log Odds Ratio

LOR.se

Standard Error for Log Odds Ratio

n.h

Degrees of freedom for log.theta

orss

Odds ratio skill score, aka Yules's Q

ORSS.se

Standard Error for Odds ratio skill score

eds

Extreme Dependency Score

esd.se

Standard Error for EDS

seds

Symmetric Extreme Dependency Score

seds.se

Standard Error for Symmetric Extreme Dependency Score

EDI

Extreme Dependency Index

EDI.se

Standard Error for EDI

SEDI

Symmetric EDI

SEDI.se

Standard Error for SEDI

Note

Initially, table.stats was an internal function used by verify for binary events and multi.cont for categorical events. But occassionally, it is nice to use it directly.

Author(s)

Matt Pocernich

References

Jolliffe, I.T. and D.B. Stephenson (2003). Forecast verification: a practitioner's guide in atmospheric science. John Wiley and Sons. See chapter 3 concerning categorical events.

Stephenson, D.B. (2000). "Use of 'Odds Ratio for Diagnosing Forecast Skill." Weather and Forecasting 15 221-232.

Hogan, R.J., O'Connor E.J. and Illingworth, 2009. "Verification of cloud-fraction forecasts." Q.J.R. Meteorol. Soc. 135, 1494-1511.

See Also

verify and multi.cont

Examples

DAT<- matrix(c(28, 23, 72, 2680 ), ncol = 2) ## Finley
table.stats(DAT)