multi.cont: Multiple Contingency Table Statistics

Description

Provides a variety of statistics for a data summarized in a contingency table. This will work for a 2 by 2 table, but is more useful for tables of greater dimensions.

Usage

multi.cont(DAT, baseline = NULL)

Arguments

DAT

A contingency table in the form of a matrix. It is assumed that columns represent observation, rows represent forecasts.

baseline

A vector indicating the baseline probabilities of each category. By default, it the baseline or naive forecasts is based on teh

Value

pcPercent correct - events along the diagonal.
biasBias
tsThreat score a.k.a. Critical success index (CSI)
hssHeidke Skill Score
pssPeirce Skill Score
gsGerrity Score
pc2Percent correct by category (vector)
hHit Rate by category (vector)
false.alarm.ratioFalse alarm ratio by category (vector)

References

Gerrity, J.P. Jr (1992). A note on Gandin and Murphy's equitable skill score. Mon. Weather Rev., 120, 2707-2712. Jolliffe, I.T. and D.B. Stephenson (2003). Forecast verification: a practitioner's guide in atmospheric science. John Wiley and Sons. See chapter 4 concerning categorical events, written by R. E. Livezey.

Examples

Run this code

DAT<- matrix(c(7,4,4,14,9,8,14,16,24), nrow = 3) # from p. 80 - Jolliffe
multi.cont(DAT)

DAT<- matrix(c(3,8,7,8,13,14,4,18,25), ncol = 3) ## Jolliffe JJA
multi.cont(DAT)

DAT<- matrix(c(50,47,54,91,2364,205,71,170,3288), ncol = 3) # Wilks p. 245
multi.cont(DAT)

DAT<- matrix(c(28, 23, 72, 2680 ), ncol = 2) ## Finley
multi.cont(DAT)
## Finnish clouds
DAT<- matrix(c(65, 10, 21, 29,17,48, 18, 10, 128), nrow = 3, ncol = 3, byrow = TRUE)
multi.cont(DAT)  
 ### alternatively, the verify function and summary can be used.
 
 mod <- verify(DAT, frcst.type = "cat", obs.type = "cat")
 summary(mod)