surveillance (version 1.12.1)

residualsCT: Extract Cox-Snell-like Residuals of a Fitted Point Process

Description

Extract the residual process (cf. Ogata, 1988) of a fitted point process model specified through the conditional intensity function, for instance a model of class "twinSIR" or "twinstim" (and also "simEpidataCS"). The residuals are defined as the fitted cumulative intensities at the event times, and are generalized residuals similar to those discussed in Cox and Snell (1968).

Usage

## S3 method for class 'twinSIR':
residuals(object, ...)
## S3 method for class 'twinstim':
residuals(object, ...)
## S3 method for class 'simEpidataCS':
residuals(object, ...)

Arguments

object

an object of one of the aforementioned model classes.

...

unused (argument of the generic).

Value

• Numeric vector of length the number of events of the corresponding point process fitted by object. This is the observed residual process.

Details

For objects of class twinstim, the residuals may already be stored in the object as component object$tau if the model was fitted with cumCIF = TRUE (and they always are for "simEpidataCS"). In this case, the residuals method just extracts these values. Otherwise, the residuals have to be calculated, which is only possible with access to the model environment, i.e. object must have been fitted with model = TRUE. The calculated residuals are then also appended to object for future use. However, if cumCIF and model were both set to true in the object fit, then it is not possible to calculate the residuals and the method returns an error.

References

Ogata, Y. (1988) Statistical models for earthquake occurrences and residual analysis for point processes. Journal of the American Statistical Association, 83, 9-27

Cox, D. R. & Snell, E. J. (1968) A general definition of residuals. Journal of the Royal Statistical Society. Series B (Methodological), 30, 248-275

See Also

checkResidualProcess to graphically check the goodness-of-fit of the underlying model.

Examples

## Load the twinSIR() fit
data("foofit")
residuals(foofit)
## these residuals are, e.g., used by checkResidualProcess()
checkResidualProcess(foofit)