MorphoRegions (version 0.1.0)

plotsegreg: Plot a segmented regression model

Description

plotsegreg() plots the fitted lines resulting from a segmented regression model.

Usage

plotsegreg(x, scores, ...)
# S3 method for regions_pco
plotsegreg(
  x,
  scores,
  modelsupport = NULL,
  criterion = "aic",
  model = 1,
  bps = NULL,
  cont = TRUE,
  ...
)
# S3 method for regions_sim
plotsegreg(
  x,
  scores,
  modelsupport = NULL,
  criterion = "aic",
  model = 1,
  bps = NULL,
  cont = TRUE,
  ...
)
# S3 method for regions_results_single
plotsegreg(x, scores, ...)

Value

A ggplot object that can be manipulated using ggplot2 syntax.

Arguments

x

a regions_pco object, the output of a call to svdPCO(), or a regions_results_single object, the output of a call to calcmodel().

scores

numeric; the indices of the PCO scores for which the fitted lines should be plotted.

...

ignored.

modelsupport

a regions_modelsupport object, the output of a call to modelsupport(). When x is a regions_pco object, either modelsupport, criterion, and model must be supplied or bps and cont must be supplied. See Details.

criterion

string; the criterion to use to select the best model for which breakpoints are to be displayed when modelsupport is specified. Ignored otherwise. Allowable options include "aic" to use the AICc and "bic" to use the BIC. Abbreviations allowed. Default is "aic". When x is a regions_pco object, either modelsupport, criterion, and model must be supplied or bps and cont must be supplied. See Details.

model

numeric; for which model among the best as determined by criterion should fitted lines be plotted. 1 is the best model, 2 the second best, etc. Default is 1. When x is a regions_pco object, either modelsupport, criterion, and model must be supplied or bps and cont must be supplied. See Details.

bps

numeric; a vector of breakpoints for which model fitted lines should be plotted. When x is a regions_pco object, either modelsupport, criterion, and model must be supplied or bps and cont must be supplied. See Details.

cont

logical; whether to fit a model that is continuous (TRUE) or discontinuous (FALSE) at the breakpoints supplied to bps. Default is TRUE. When x is a regions_pco object, either modelsupport, criterion, and model must be supplied or bps and cont must be supplied. See Details.

Details

plotsegreg() operates on a single model identified by breakpoints and whether the model is continuous or not. When x is a regions_pco object, the model is selected either as the best model in the supplied modelsupport object (where "best" is determined by the arguments to criterion and model) or as specified by the user using the arguments to bps and cont. When x is a regions_results_single object, the breakpoints and model form are determined based on the supplied object.

plot() is an alias for plotsegreg() for regions_results_single objects.

See Also

modelsupport() for assessing model support using information criteria; calcmodel() for fitting a single segmented regression model; modelperf() for computing fit statistics for a single segmented regression model.

Examples

data("alligator")

alligator_data <- process_measurements(alligator,
                                       pos = "Vertebra")

# Compute PCOs
alligator_PCO <- svdPCO(alligator_data)

# Evaluate model performance (R2) given supplied
# breakpoints for a continuous model
modelperf(alligator_PCO, scores = 1:3,
          bps = c(7, 15, 20), cont = TRUE)

plotsegreg(alligator_PCO, scores = 1:3,
           bps = c(7, 15, 20), cont = TRUE)
## See also `?calcmodel` for use with a single model

# Fit segmented regression models for 1 to 7 regions
# using PCOs 1 to 4 and a continuous model with a
# non-exhaustive search
regionresults <- calcregions(alligator_PCO,
                             scores = 1:4,
                             noregions = 7,
                             minvert = 3,
                             cont = TRUE,
                             exhaus = FALSE,
                             verbose = FALSE)

regionresults

# For each number of regions, identify best
# model based on minimizing RSS
bestresults <- modelselect(regionresults)

# Evaluate support for each model and rank
supp <- modelsupport(bestresults)

# Evaluate model performance (R2) for best model
# as chosen by BIC
modelperf(alligator_PCO, scores = 1:4,
          modelsupport = supp,
          criterion = "bic", model = 1)

# Plot that model for the first PCO score
plotsegreg(alligator_PCO, scores = 1:4,
           modelsupport = supp,
           criterion = "bic", model = 1)

## See `?simregions` for use with simulated data