calcGwithK: calcGwithK - Calculate eoa's g value
In tmcd82070/EoAR: Evidence of Absence Regression (EoAR)

Description Usage Arguments Details Value Author(s) Examples

Compute EoA g = Pr(discovery) given searcher efficiency, carcass removal distribution, area correction, "k", and search dates.

1	calcGwithK(searchDates, crShape, crScale, crDist, seef, k, a)

`searchDates`	A vector of search dates. This can either be a character vector containing representations of real dates (e.g., c('2014-03-01','2014-04-01', etc.)), a vector of `POSIXct` objects representing real dates, or a vector of integers specifying days after start of the study (e.g., c(0,30,60, etc.)). If `searchDates` is a vector of integers, the first element must be 0. If `searchDates` is a vector of characters, format of the dates must be " (or YYYY-MM-DD).
`crShape`	Shape parameter of the persistence distribution for objects. If `crDist` == "exponential", this parameter is ignored.
`crScale`	Scale parameter of the persistence distribution for objects. If `crDist` == "exponential", this parameter is the mean, which is 1/rate, where rate is the parameter of `pexp` in R.
`crDist`	Character string naming the persistence distribution to use. Valid values are "exponential", "weibull", "lognormal", and "loglogistic". Anything else throws an error.
`seef`	A scalar specifying probability of detection. This is probability of detecting an object given it is present.
`k`	A scalar specifying the proportion of objects remaining available for detection following the first search they became available. E.g., `k` of the objects missed on a search are available during the next search.
`a`	A scalar specifying the proportion of the object's spatial distribution searched, sometimes called area correction proportion.

This routine assumes the arrival function of objects on the study area is uniform between min(searchDates) and max(searchDates).

A scalar, the probability of object discovery, which is the product of probability of an object being available and probability of detection given presence.

Trent McDonald, with help from Jared Studyvin, Kristen Nasmen, and Paul Rabie. Based largely on code in package eoa by Dan Dalthorp.

days <- c(0,7,14,21,28,58,88)
calcGwithK(days, NA, 21, "exponential", 0.7, 0.2, 0.5)

days <- c("2014-05-15","2014-06-15","2014-08-15")
calcGwithK(days, NA, 21, "exponential", 0.7, 0.2, 0.5)

startDay <- c("2014-05-15")
endDay <- c("2014-09-15")
searchInterval <- 7 # days
nSearches <- floor(as.numeric(as.POSIXct(endDay) - as.POSIXct(startDay))/searchInterval)
days <- as.POSIXct(startDay) + (0:nSearches)*searchInterval*(60*60*24)
calcGwithK(days, 0.57792, 7.16, "weibull", 0.7, 0.2, 0.5)