qqnorm.STdata: QQ-norm for 'STdata'/'STmodel'/'predCVSTmodel' objects

In SpatioTemporal: Spatio-Temporal Model Estimation

Description

qqnorm method for classes STdata/STmodel/predCVSTmodel. Used for data and residual analysis of the cross validation.

Usage

## S3 method for class 'predCVSTmodel'
qqnorm(y, ID = "all",
  main = "Q-Q plot for CV residuals", group = NULL, col = 1,
  norm = FALSE, line = 0, org.scale = TRUE, ...)

## S3 method for class 'STdata'
qqnorm(y, ID = "all", main = "Q-Q plot for observations",
  group = NULL, col = 1, line = 0, ...)

## S3 method for class 'STmodel'
qqnorm(y, ID = "all", main = "Q-Q plot for observations",
  group = NULL, col = 1, line = 0, ...)

Arguments

y	STdata/STmodel/predCVSTmodel object for the qqnorm.
ID	The location for which we want to norm-plot observations/residuals or "all" to plot for all locations.
main	Title of the plot
group	Do the norm-plot both for all data and then for each subset defined by the factor/levels in group variable.
col	Colour of points in the plot, either a scalar or a vector with length matching the number of observations/residuals.
norm	TRUE/FALSE, plot normalised (mean=0, sd=1) or raw cross-validation residuals. If norm=TRUE a 0-1 line is added, to indicate what normalised residuals should look like.
line	If non-zero add a qqline with lty=line, to the plot; if 0 do not add a line.
org.scale	TRUE/FALSE scatter plots on the original untransformed scale, or using exp(y). Only relevant if x was computed using transform in predictCV.STmodel (as pass through argument to predict.STmodel)
...	Arguments passed on to the plotting function, qqnorm.

Value

Nothing

Author(s)

Johan Lindstrom

See Also

Other predCVSTmodel methods: estimateCV.STmodel, plot.predCVSTmodel, print.predCVSTmodel, print.summary.predCVSTmodel, scatterPlot.predCVSTmodel, summary.predCVSTmodel

Other STdata methods: createSTdata, plot.STdata, print.STdata, print.summary.STdata, scatterPlot.predCVSTmodel, summary.STdata

Other STmodel methods: MCMC.STmodel, c.STmodel, createSTmodel, estimate.STmodel, estimateCV.STmodel, plot.STdata, predict.STmodel, print.STmodel, print.summary.STmodel, scatterPlot.predCVSTmodel, simulate.STmodel, summary.STmodel

Examples

################################
## Example for STdata/STmodel ##
################################
##load data
data(mesa.model)

##standard plot
qqnorm(mesa.model)
##add a line, and group (and colour) by AQS/FIXED
par(mfrow=c(2,2))
obs.type <- mesa.model$locations$type[match(mesa.model$obs$ID,
                                            mesa.model$locations$ID)]
qqnorm(mesa.model, line=1, group=obs.type, col=obs.type)

##colour code by season and split by type
##First create a vector dividing data into four seasons
I.season <- as.factor(as.POSIXlt(mesa.model$obs$date)$mon+1)
levels(I.season) <- c(rep("Winter",2), rep("Spring",3), 
                      rep("Summer",3), rep("Fall",3), "Winter") 

par(mfrow=c(2,2))
qqnorm(mesa.model, line=1, col=I.season, group=obs.type)
legend("bottomright", legend=as.character(levels(I.season)),
       pch=1, col=1:nlevels(I.season))

###############################
## Example for predCVSTmodel ##
###############################
##load data
data(pred.cv.mesa)

##standard plot
par(mfrow=c(1,1))
qqnorm(pred.cv.mesa, line=3)
##or for the normalised residuals
qqnorm(pred.cv.mesa, line=3, norm=TRUE)

##add a line, and group by AQS/FIXED
par(mfrow=c(2,2))
qqnorm(pred.cv.mesa, line=1, group=obs.type)

##and for normalised residuals, colour-coded by season
par(mfrow=c(2,2))
qqnorm(pred.cv.mesa, line=2, norm=TRUE,
       group=obs.type, col=I.season)
legend("bottomright", legend=as.character(levels(I.season)),
       pch=1, col=1:nlevels(I.season))

SpatioTemporal documentation built on May 2, 2019, 8:49 a.m.