lagfit: Fits a piecewise glm model with lags

In PopulationGrowthR: Linear Population Growth Scenarios

Fits a piecewise glm model with lags

Description

This function fits a piecewise poisson model to the frequency data of different Species. It assumes that the data contains columns Year, Frequency and Specimens.

Usage

lagfit(
  data,
  yeardata,
  species = NULL,
  knots = NULL,
  zeros = TRUE,
  plotlag = FALSE,
  plotfreq = FALSE
)

Arguments

data	a dataframe containing the columns Species (optional), Year, Frequency and Specimens.
yeardata	a dataframe containing the columns Year and Specimens giving the total number of Specimens for each Year.
species	list of species for which the model is to be fitted. Default is NULL, which fits the model for all species in the data.
knots	a list of knots to be used for the piecewise model. Default is NULL, which chooses the optimal model with 0-4 knots.
zeros	logical. Specifies whether missing year for the species will be filled with zeros. Default is TRUE.
plotlag	logical. If TRUE a plot of the fitted model will be produced for each species.
plotfreq	logical. If TRUE frquency plots will be created for each species.

Value

If the model is fit for a single species following are returned as a list

• Species - Species name

• Scene - Different scenario of the fit between the knots. A sequence of 0, + or - is returned. A 0 indicates constant, + indicates increasing and a - indicates decreasing.

• Lag - Logical. Is there a lag present or not.

• Laglength - Length of the first lag. Position of the First Knot - the first year for that species

• FirstYear - The first year for that species for which data is available.

• EndYear - The first knot position.

• fit - the fitted model.

If the number of species is more than one, then a list is returned with following items:

• fitdata - dataframe is returned with the items in the above list except for the fitted model.

• fitcoefs - list of coefficients for the piecewise fits for each Species

Examples

#Run lagfit for 1 species only
Species = unique(fdata$Species) #List of all species

fit1 = lagfit(fdata, yeardata, species=Species[1])
#Run lagfit for multiple species
fit2 = lagfit(fdata, yeardata, species=Species[1:3])
fitdata = fit2$fitdata  #Dataframe containing fits
fitcoefs = fit2$fitcoefs #List containing slopes of the fitted splines

## Not run: 
#Run lagfit for the whole dataset
fitall = lagfit(fdata, yeardata)

## End(Not run)

PopulationGrowthR documentation built on May 10, 2022, 5:11 p.m.