lmf: Fitting age-structured selection model

Description

lmf fit linear models within each combination of year and age class and estimates coefficients of selection using maximum likelihood procedures. lmf is compatible with populations without age-structure.

Usage

lmf(formula, age, year, data, na.action = na.exclude,
method = c("BFGS"), control = list(maxit = 500,
reltol = sqrt(.Machine$double.eps)), ...)

Arguments

formula

an object of class "formula" (or one that can be coerced to that class): a symbolic description of the model to be fitted when estimating coefficients of selection. Format: response ~ terms. The detail of model specification are given under 'Details'.

age

used to define the name of the age column in the data set. Use NULL if no age data are available in the data set.

year

used to define the name of the year column in the data set.

data

data set with individual ids (optional), year of reproduction (year), maternal age (age; may be omitted if a model without age is desired), number of female offspring (recruits), survival to the next reproductive event (survival) and phenotypic measurements. Age classes should have a natural order of increasing age. E.g. 1, 2, 3, ...

na.action

a function which indicate what should happend when the data contain NAs. The default is na.exclude (see ?na.fail).

method

defines what optimalization algorithm to be used in the maximization of the loglikelihood. Alternatives are: "Nelder-Mead", "BFGS" (default), "CG", "L-BFGS-B" and "SANN". Not all are applicable here. See ?optim for details.

control

a list of control parameters for the maximization of the likelihood. maxit sets the maximum number of iterations to use before convergence and reltol sets the relative threshold for improvement in the likelihood which desides whether to continue maximation or end. See ?optim for details.

...

additional arguments to be passed to optim for the maximization of the loglikelihood. See ?optim for options.

Value

running.time: the total time used for computation.
optim.time: the time used for maximation of the loglikelihood.
call: the matched call.
npar: the number of parameters in the model.
uage: the unique age classes in the data set.
nage: the number of unique age classes in the data set.
maxage: the final age class.
l: the estimated projection matrix.
lambda: the deterministic multiplicative growth rate of the population.
u: the stable age distribution.
v: the vector of reproductive values for each age class.
uyear: the unique years in the data set.
nyear: the number of unique years in the data set.
nobs: the number of observations (counting individual-year).
nobs.age: the number of observations per age class.
indnr: assigned individual numbers (1:nobs).
ajt: a list containing the named vectors of the estimated selection coefficient for each age class within each year. Sorted by age class and year.
Ajt.us: a list containing the named unscaled variance-covariance matrix for each age class within each year. Sorted by age class and year.
sigma.djt: a list containing the vectors of residual standard errors from the linear regression for each age class within each year. Sorted by age class and year.
dof: a list containing the vectors of degrees of freedom (dof) from the linear regression for each age class within each year. Sorted by age class and year.
res: a list containing the vectors of residuals from the linear regression for each age class within each year. Sorted by age class and year.
fit: a list containing the vectors of fitted values from the linear regression for each age class within each year. Sorted by age class and year.
leverage: a list containing the vectors of estimated leverage for each data point from the linear regression for each age class within each year (see lm.influence?). Sorted by age class and year.
cook: a list containing the vectors of estimated Cook's distance for each data point from the linear regression for each age class within each year (see cooks.distance?). Sorted by age class and year.
sigma2.dj: a list containing the demographic variance for each age class. Sorted by age class.
sigma2.dj.dof: a list containing the degrees of freedom (dof) for the demographic variance for each age class. Sorted by age class.
sigma2.dj.sd: a list containing the standard deviation (sd) for the demographic variance for each age class. Sorted by age class.
sigma2.d: the total demographic variance of the population.
sigma2.d.dof: the degrees of freedom (dof) for the total demographic variance of the population.
sigma2.d.sd: the standard deviation (sd) for the total demographic variance of the population.
Ajt: a list containing the named variance-covariance matrix (scaled by sigma2.dj) for each age class within each year. Sorted by age class and year.
at: a list containing the named vectors of the estimated selection coefficient for each year. Sorted by year.
At: a list containing the named variance-covariance matrix (scaled by sigma2.dj) for each year. Sorted by year.
convergence: "yes" indicates that the numerical maximation of the likelihood successfully converged before reaching the iteration limit maxit.
iterations: the number of iterations of the function in the numerical maximation of the likelihood.
M: the estimated temporal covariance matrix (fluctuating selection).
aM: the estimated temporal mean selection coefficients.
atC: the best linear predictor for the estimated yearly selection coefficients (i.e. corrected for sampling errors).
Anf: the estimated temporal covariance matrix assuming no fluctuating selection.
anf: the estimated temporal mean selection coefficients assuming no fluctuating selection.
sigma2.e: the environmental variance of the population.
data.set: the data set used in the analyses with a column of individual reproductive values addad

Details

lmf use formulas for model specification. These should be formatted as decribed under arguments. Note however that your response should be specified as a two-column matrix with the columns recruits and survival. The first column should give the number of recruits that an individual produced a given year and the second column should contain information on whether the individual survived or not (1 or 0) to the next breeding season. These two columns will be used to calculate the individual reproductive values (Wj) which the model will substitute for the response in the age and year specific linear regressions (i.e. Wj ~ terms).

References

Engen, S., Saether, B.-E., Kvalnes, T. and Jensen, H. 2012. Estimating fluctuating selection in age-structured populations. Journal of Evolutionary Biology, 25, 1487-1499.

Examples

Run this code

#Data set from Engen et al. 2012
data(sparrowdata)
#Fit model
lmf.1 <- lmf(formula = cbind(recruits, survival) ~ weight + tars,
               age = age, year = year, data = sparrowdata)
#View diagnostic plots
plot(lmf.1)
#View output
print(lmf.1)
#Print summary
summary(lmf.1)

Run the code above in your browser using DataLab