BCSreg (version 1.1.1)

plot.BCSreg: Diagnostic Plots for a Box-Cox Symmetric Regression Fit

Description

This function provides plots for diagnostic analysis of a Box-Cox symmetric or a zero-adjusted regression fit.

Usage

# S3 method for BCSreg
plot(
  x,
  which = 1:4,
  ask = prod(graphics::par("mfcol")) < length(which) && grDevices::dev.interactive(),
  pch = "+",
  las = 1,
  cex = 0.8,
  lwd = 2,
  ...
)

Value

plot method for "BCSreg" objects returns seven types of diagnostic plots.

Arguments

x

an object of class "BCSreg".

which

numeric; if a subset of the plots is required, specify a subset of the numbers 1:7.

ask

logical; if TRUE, the user is asked before each plot.

pch, las, cex, lwd, ...

graphical parameters (see par)

Author

Francisco F. de Queiroz <>

Rodrigo M. R. de Medeiros <>

Details

The plot method for BCSreg objects provides seven types of diagnostic plots in the following order:

Residuals vs fitted values

a plot of the residuals versus the fitted medians.

Residuals vs observation indices.

an index plot of the residuals versus the observation indices.

Density plot

a graph that compares the empirical density of the residuals with the density of the standard normal distribution.

Normal probability plot

a normal probability plot of the residuals with a confidence region constructed according to Fox (2016) using the qqPlot function.

Case-weight perturbation

An index plot of local influence based on the case-weight perturbation scheme.

Fitted vs observed values

a dispersion diagram of the fitted values versus the observed values.

Residuals vs v(z) function

a dispersion diagram of the \(v(z)\) function versus the residuals. For some BCS models, the \(v(z)\) function may be interpreted as weights in the estimation process. If family = "NO", the \(v(z)\) function is constant.

The which argument can be used to select a subset of the implemented plots. Default is which = 1:4. See residuals.BCSreg for details on the residuals.

References

Medeiros, R. M. R., and Queiroz, F. F. (2025). Flexible modeling of nonnegative continuous data: Box-Cox symmetric regression and its zero-adjusted extension.

Examples

## Data set: raycatch (for description, run ?raycatch)
hist(raycatch$cpue, xlab = "Catch per unit effort")
plot(cpue ~ tide_phase, raycatch, pch = 16,
    xlab = "Tide phase", ylab = "Catch per unit effort")
plot(cpue ~ location, raycatch, pch = 16,
    xlab = "Location", ylab = "Catch per unit effort")
plot(cpue ~ max_temp, raycatch, pch = 16,
    xlab = "Maximum temperature", ylab = "Catch per unit effort")

## Fit a double Box-Cox normal regression model:
fit <- BCSreg(cpue ~ location + tide_phase |
                location + tide_phase + max_temp, raycatch)

## Available plots:

### Residuals vs fitted values (fitted medians)
plot(fit, which = 1)

### Residuals vs observation indices
plot(fit, which = 2)

### Density plot
plot(fit, which = 3)

### Normal probability plot
plot(fit, which = 4)

### Local influence
plot(fit, which = 5)

### Fitted medians vs response
plot(fit, which = 6)

### v(z) function
plot(fit, which = 7)