CODEWITHSUNDEEP
r
Gretchen Newberry
Gretchen Newberry

Reputation: 1

Using the Dredge function in RMARK

I'm using RMark to calculate nest success vs. 8 covariates.

My data looks like:

1) Year
2) Julian Day Egg Lay Date
3) Proportion of Eggs Layed
4) Of eggs Layed, Proportion of Eggs Hatched
5) Mean ibutton Temp Before Hatching
6) Max iButton Temp Before Hatching
7) Mean Te Before Hatching
8) Max Te Before Hatching

Nest Survival Group=1 ; 32 2 8 8 0 11 2016 152 1 0.5 30.56 60.5 33.46 71.11 ;

I'm trying to use the dredge function to do all possible combinations.

My code to do single predictor vs. response varable:

library(RMark)
hatch = scan("C:.../mark file hatch Julian-150forR.inp", what = "character", sep = "\n")  #to find the file
write(sub(";", "", hatch[13:56]), "hatch.txt")
hatch = read.table("hatch.txt")
names(hatch) = c("id", "FirstFound", "LastPresent", "LastChecked", "Fate", "Freq", 
    "Year", "LayDate", "PropLayed", "PropHatch", "MeaniButtn", "MaxiButtn", "MeanTe", 
    "MaxTe")
run.hatch = function() {
    Dot = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~1)))
    Year = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~Year)))
    LayDate = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~LayDate)))
    PropLayed = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~PropLayed)))
    MeaniButtn = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~MeaniButtn)))
    MaxiButtn = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~MaxiButtn)))
    MeaniTe = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~MeanTe)))
    MaxTe = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~MaxTe)))
    return(collect.models())
}
hatch.results = run.hatch()

When I try to use the dredge function, here's my code:

require(MuMIn)
run.hatch = function() {
    global <- Dot = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~1))) + 
        Year = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~Year))) + 
        LayDate = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~LayDate))) + 
        PropLayed = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~PropLayed))) + 
        MeaniButtn = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~MeaniButtn))) + 
        MaxiButtn = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~MaxiButtn))) + 
        MeaniTe = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~MeanTe))) + 
        MaxTe = mark(hatch, nocc = 58, model = "Nest", model.parameters = list(S = list(formula = ~MaxTe))) + 
        return(collect.models())
}
combinations <- dredge(global)


> Error in nobs(global.model) : object 'global' not found

Any suggestions?

Upvotes: 0

Views: 231

Answers (1)

Kamil Bartoń
Kamil Bartoń

Reputation: 1562

You define global within the scope of function run.hatch. It is invisible outside of it. Besides, run.hatch is supposed to return a list of models, while dredge expects just one model. For example:

global <- mark(...)
combinations <- dredge(global)

Upvotes: 1

Related Questions