itsadug (version 2.4.1)

plot_modelfit: Visualization of the model fit for time series data.

Description

Plots the fitted values and the data for n trials of time series data. For example, plots n trials of the same participant.

Usage

plot_modelfit(
  x,
  view,
  event = NULL,
  n = 3,
  random = TRUE,
  cond = NULL,
  col = c(alpha(1), "red"),
  add = FALSE,
  eegAxis = FALSE,
  fill = FALSE,
  main = NULL,
  xlab = NULL,
  ylab = NULL,
  ylim = NULL,
  h0 = 0,
  v0 = NULL,
  transform = NULL,
  hide.label = FALSE,
  hide.legend = FALSE,
  print.summary = getOption("itsadug_print"),
  ...
)

Arguments

x

A lm or gam object, produced by gam or bam, lm, glm.

view

Text string containing the predictor or column in the data to be displayed on the x-axis. Note that variables coerced to factors in the model formula won't work as view variables.

event

column name from the data that specifies the time series from which n are being plotted.

n

Number of time series to plot. Default is 3. Set to -1 for plotting all time series (which may take a considerable time).

random

Numeric: if set to TRUE (default), n random events are selected to plot. If set to FALSE, the first n events are selected to plot. The events could be precisely controlled with the argument cond.

cond

A named list of the values to use for the other predictor terms (not in view) or to select specific trials or time series to plot.

col

Two value vector specifiying the colors for the data and the modelfit respectively.

add

Logical: whether or not to add the lines to an existing plot, or start a new plot (default).

eegAxis

Logical: whether or not to reverse the y-axis, plotting the negative amplitudes upwards as traditionally is done in EEG research. If eeg.axes is TRUE, labels for x- and y-axis are provided, when not provided by the user. Default value is FALSE.

fill

Logical: whether or not to fill the area between the data and the fitted values with shading. Default is FALSE.

main

Changing the main title for the plot, see also title.

xlab

Changing the label for the x axis, defaults to a description of x.

ylab

Changing the label for the y axis, defaults to a description of y.

ylim

the y limits of the plot.

h0

A vector indicating where to add solid horizontal lines for reference. By default no values provided.

v0

A vector indicating where to add dotted vertical lines for reference. By default no values provided.

transform

Function for transforming the fitted values. Default is NULL.

hide.label

Logical: whether or not to hide the label (i.e., 'fitted values'). Default is FALSE.

hide.legend

Logical: whether or not to hide the legend. Default is FALSE.

print.summary

Logical: whether or not to print a summary. Default set to the print info messages option (see infoMessages).

...

other options to pass on to lines and plot, see par

Notes

This function plots the fitted effects, including intercept and other predictors.

See Also

Other Model evaluation: check_resid(), diagnostics()

Examples

# NOT RUN {
data(simdat)

# Create grouping predictor for time series:
simdat$Event <- interaction(simdat$Subject, simdat$Trial)

# model without random effects:
m1 <- bam(Y ~ te(Time, Trial),
    data=simdat)
plot_modelfit(m1, view='Time', event=simdat$Event)

# colorizing residuals:
plot_modelfit(m1, view='Time', event=simdat$Event, fill=TRUE)

# All trials of one subject:
# }
# NOT RUN {
# this produces error:
plot_modelfit(m1, view='Time', event=simdat$Event, 
    cond=list(Subject='a01'), n=-1)
# }
# NOT RUN {
# instead try this:
simdat$Subj <- ifelse(simdat$Subject=='a01', TRUE, FALSE)
plot_modelfit(m1, view='Time', event=simdat$Event, 
    cond=list(Subject=simdat$Subj), n=-1)

# }
# NOT RUN {
# Model with random intercepts for subjects:
m2 <- bam(Y ~ te(Time, Trial)+s(Subject, bs='re'),
    data=simdat)
# now selecting a subject works, because it is in the model:
plot_modelfit(m2, view='Time', event=simdat$Event, 
    cond=list(Subject='a01'), n=-1, ylim=c(-13,13))

# Model with random effect and interactions:
m3 <- bam(Y ~ te(Time, Trial)+s(Time, Subject, bs='fs', m=1),
    data=simdat)
plot_modelfit(m3, view='Time', event=simdat$Event, 
    cond=list(Subject='a01'), n=-1, ylim=c(-13,13))
# }