GOFPlot: Goodness of fit pp-plot

In dpwynne/mmnst: Multiscale Modeling of Neural Spike Trains

Goodness of fit pp-plot

Description

Runs a goodness-of-fit test for a particular model, i.e., a given intensity function estimate.

Usage

GOFPlot(
  spikes,
  theta,
  t.start = 0,
  t.end = 10,
  neuron.name = NULL,
  resolution = (t.end - t.start)/(nrow(theta) - 1),
  axis.label.size = 18,
  title.size = 24
)

Arguments

spikes	a list of spike trains.
theta	a numeric matrix in which the jth column contains the intensity function estimate for the j spike train in the list.
t.start	the starting time of the recording window; the default value is 0.
t.end	the ending time of the recording window. The default value is 10, corresponding to a 10-second recording.
neuron.name	a string containing the name of the neuron being plotted, used only to title the plot.
resolution	a scalar determining the bin width. This is equivalent to Delta in Haslinger et al. (2010).
axis.label.size	a scalar determining the font size of the x- and y-axis labels of the plot.
title.size	a scalar determining the font size of the title of the plot (if one exists).

Value

a list of data frames. Each data frame includes the model and empirical quantiles for one spike train. This data is used to generate the pp-plot outlined in Haslinger et al. (2010).

References

Haslinger, R., Pipa G., and Brown, E. (2010). Discrete time rescaling theorem: determining goodness of fit for discrete time statistical models of neural spiking. Neural Computation. 22(10):2477-506. doi: 10.1162/NECO_a_00015.

dpwynne/mmnst documentation built on Aug. 1, 2023, 8:08 a.m.