multi.cont {verification} | R Documentation |
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.
multi.cont(DAT)
DAT |
A contingency table in the form of a matrix. It is assumed that columns represent observation, rows represent forecasts. |
pc |
Percent correct - events along the diagonal. |
bias |
Bias |
pod |
Probability of detecting an event. |
far |
False Alarm Rate |
ts |
Threat score a.k.a. Critical success index (CSI) |
hss |
Heidke Skill Score |
pss |
Pierce Skill Score |
gs |
Gerrity Score |
Matt Pocernich <pocernic@rap.ucar.edu>
Gerrity, J.P. Jr (1992). A note on Gandin and Murphy's equitable skill score. Mon. Weather Rev., 120, 2707-2712.
Joliffe, 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.
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)