RMark (version 3.0.0)

killdeer: Killdeer nest survival example data

Description

A data set on killdeer that accompanies MARK as an example analysis for the nest survival model.

Format

A data frame with 18 observations on the following 6 variables.

id

a MARK comment field with a nest id

FirstFound

the day the nest was first found

LastPresent

the last day that chicks were present

LastChecked

the last day the nest was checked

Fate

the fate of the nest; 0=hatch and 1 depredated

Freq

the frequency of nests with this data; usually 1

Details

This is a data set that accompanies program MARK as an example for nest survival. The data structure for the nest survival model is completely different from the capture history structure used for most MARK models. To cope with these data you must import them into a dataframe using R commands and assign the specific variable names shown above. The id and Freq fields are optional. Freq is assumed to be 1 if not given. You cannot import the MARK .inp file structure directly into R without some manipulation. Also note that import.chdata and convert.inp do NOT work for nest survival data. In the examples section below, the first section of code provides an example of converting the killdeer.inp file into a dataframe for RMark.

If your dataframe contains a variable AgeDay1, which is the age of the nest on the first occasion then you can use a variable called NestAge which will create a set of time-dependent covariates named NestAge1,NestAge2 ...NestAge(nocc-1) which will provide a way to incorporate the age of the nest in the model. This was added because the age covariate in the design data for S assumes all nests are the same age and is not particularly useful. This effect could be incorporated by using the add() function in the design matrix but RMark does not have any capability for doing that and it is easier to create a time-dependent covariate to do the same thing.

Examples

# \donttest{
# This example is excluded from testing to reduce package check time
# EXAMPLE CODE FOR CONVERSION OF .INP TO NECESSARY DATA STRUCTURE
# read in killdeer.inp file
#killdeer=scan("killdeer.inp",what="character",sep="\n")
# strip out ; and write out all but first 2 lines which contain comments
#write(sub(";","",killdeer[3:20]),"killdeer.txt")
# read in as a dataframe and assign names
#killdeer=read.table("killdeer.txt")
#names(killdeer)=c("id","FirstFound","LastPresent","LastChecked","Fate","Freq")
#
# EXAMPLE CODE TO RUN MODELS CONTAINED IN THE MARK KILLDEER.DBF
data(killdeer)
# produce summary
summary(killdeer)
# Define function to run models that are in killdeer.dbf
# You must specify either the number of occasions (nocc) or the time.intervals 
# between the occasions.
run.killdeer=function()
{
   Sdot=mark(killdeer,model="Nest",nocc=40,delete=TRUE)
   STime=mark(killdeer,model="Nest",
       model.parameters=list(S=list(formula=~I(Time+1))),nocc=40,threads=2,delete=TRUE)
   STimesq=mark(killdeer,model="Nest",
       model.parameters=list(S=list(formula=~I(Time+1)+I((Time+1)^2))),nocc=40,threads=2,
            delete=TRUE)
   STime3=mark(killdeer,model="Nest",
      model.parameters=list(S=list(formula=~I(Time+1)+I((Time+1)^2)+I((Time+1)^3))),
                   nocc=40,threads=2,delete=TRUE)
   return(collect.models())
}
# run defined models
killdeer.results=run.killdeer()
# }