makeFecObj: Function to build fecundity objects

Description

Allows a series of different glms to be fit all on the way to fecundity, e.g., probability of flowering, number of flower heads produced, etc; as well as fecundity into different discrete classes.

Usage

makeFecObj(dataf, fecConstants=data.frame(NA),
		Formula=list(fec~size),Family="gaussian",
		Transform="none",meanOffspringSize=NA,
		sdOffspringSize=NA,offspringSplitter=data.frame(continuous=1),
		vitalRatesPerOffspringType=data.frame(NA),fecByDiscrete=data.frame(NA),
		offspringSizeExplanatoryVariables="1")

Arguments

dataf

a dataframe with columns "size", "sizeNext", "stage", "stageNext", and any additional columns with fecundity data. UIf fecundity data is transformed via log, etc, this MUST BE MADE CLEAR in the argument Transform since the fecundity object pr

fecConstants

a list containing the value by which each of the fecundity rates will be multiplied in the order defined by the order in Formula. This data frame adjusts the probability of establishment of seeds or other stages in sexual reproduct

Formula

a formulas describing the desired explanatory variables (interactions, etc) in classical R style, i.e. separated by `+', `*', `:' and the response variables of choice. Possible covariates include `size', 'size2' (size^2), `size3' (size^3),`logsize' (log(

Family

a character vector containing the names of the families to be used for the glms, e.g., binomial, poisson, etc. Again, these must appear in the order defined by Formula

Transform

a character vector containing the names of the transforms to be used for the response variables, e.g., log, sqrt, -1, etc. Again, these must appear in the order defined by Formula

meanOffspringSize

numeric vector, defining mean offspring size. Defaults to NA, in which case the function will use to data to assess the mean offspring size according to the relationship defined in offspringSizeExplanatoryVariables (which either simply fits a mean, or may

sdOffspringSize

numeric vector, defining standard deviation of offspring size. Defaults to NA, in which case the function will use the data to assess the standard deviation of offspring size; as described for meanOffspringSize

offspringSplitter

dataframe with values defining the number of offspring going into the indicated offspring category; will be re-scaled to sum to 1 within the function. This argument needs to be entered as a data.frame, and the names in the data.frame need to preci

vitalRatesPerOffspringType

dataframe defining which fecundity rates (both functions and constants) apply to which offspring category. This only needs to be specified when some fecundity rates do not apply to all offspring categories. The offspring categories in the column names of

fecByDiscrete

data.frame defining number of offspring produced by each discrete class ; defaults to 0. If specified, ALL discrete classes MUST appear in alphabetical order, so NO "continuous". e.g. fecByDiscrete=data.frame(dormant=0,seedAge1=4.2,seedOld=0)

offspringSizeExplanatoryVariables

a character defining the relationship defining offspring size; the default is "1", indicating simply fitting a mean and a variance; alternatives would including defining offspring size as a function of maternal size (i.e., offspringSizeExplanatoryVariable

Value

an object of class fecObj

Details

This function fits a suite of subfunctions of fecundity towards creating a Fecundity transition projection model; e.g., the probability of flowering as a function of size, the number of seeds produced as a function of size, etc. Users can define the functional form of each relationship, as well as the distribution and any transforms. There is also a possibility of defining reproduction from discrete sizes into each of the subfunction outcomes; defined in the matrix fecByDiscrete. Note that it is crucial that users appropriately set up the data to adequately reflect conditionality in the fertility kernel; for example, if there are two columns, with one reflecting the probability of flowering (0s and 1s) and the other reflecting seed output (integers) it is important that where the probability of flowering is 0, seed output is set to NA, as otherwise, meaningless 0s in the seed output column will bias the regression. Finally, if the same variables are used in multiple equations, users should check for problems of endogeneity.

Examples

Run this code

dff <- generateDataDiscrete()
fv1 <- makeFecObj(dff, offspringSplitter = data.frame(continuous = 0.2, seedAge1 = 0.8),
	fecByDiscrete = data.frame(dormant = 0, seedAge1 = 4.2, seedOld = 0))