bodaDelay: Bayesian aberration detection in presence of reporting...

In jimhester/surveillance: Temporal and Spatio-Temporal Modeling and Monitoring of Epidemic Phenomena

Description

The function takes range values of the surveillance time series sts and for each time point uses a Bayesian model of the negative binomial family with log link inspired by the work of Noufaily et al. (2012) and of Manitz and Höhle (2014). It allows delay-corrected aberration detection as explained in Salmon et al. (2015). A reportingTriangle has to be provided in the control slot.

Usage

bodaDelay(sts, control = list(range = NULL, b = 3, w = 3,
                                          mc.munu=100, mc.y=10,
                                          pastAberrations = FALSE,
                                          verbose = FALSE,
                                          alpha = 0.01, trend = TRUE,
                                          limit54=c(5,4),
                                          inferenceMethod=c("asym","INLA"),
                                          noPeriods = 1, pastWeeksNotIncluded = 26,
                                          delay = TRUE))

Arguments

sts	sts-object to be analysed. Needs to have a reporting triangle.
control	list with control arguments
b	How many years back in time to include when forming the base counts.
w	Window's half-size, i.e. number of weeks to include before and after the current week in each year.
range	Specifies the index of all timepoints which should be tested. If range is NULL all possible timepoints are used.
pastAberrations	Boolean indicating whether to include an effect for past outbreaks in a second fit of the model. This option only makes sense if inferenceMethod is INLA, as it is not supported by the other inference method.
verbose	Boolean specifying whether to show extra debugging information.
alpha	An approximate (one-sided) (1-α)\cdot 100\% prediction interval is calculated unlike the original method where it was a two-sided interval. The upper limit of this interval i.e. the (1-α)\cdot 100\% quantile serves as an upperbound.
trend	Boolean indicating whether a trend should be included
noPeriods	Number of levels in the factor allowing to use more baseline. If equal to 1 no factor variable is created, the set of reference values is defined as in Farrington et al (1996).
inferenceMethod	Which inference method used, as defined in Salmon et al (2015). If one chooses INLA then inference is performed with INLA. If one chooses asym then the asymptotic normal approximation of the posteriori is used.
pastWeeksNotIncluded	Number of past weeks to ignore in the calculation.
delay	Boolean indicating whether to take reporting delays into account.
mc.munu	Number of samples for the parameters of the negative binomial distribution when performing Monte Carlo to calculate a threshold
mc.y	Number of samples for observations when performing Monte Carlo to calculate a threshold
limit54	c(cases,period) is a vector allowing the user to change these numbers.

References

A statistical algorithm for the early detection of outbreaks of infectious disease, Farrington, C.P., Andrews, N.J, Beale A.D. and Catchpole, M.A. (1996), J. R. Statist. Soc. A, 159, 547-563. An improved algorithm for outbreak detection in multiple surveillance systems, Noufaily, A., Enki, D.G., Farrington, C.P., Garthwaite, P., Andrews, N.J., Charlett, A. (2012), Statistics in Medicine, published online. Bayesian outbreak detection in the presence of reporting delays, Salmon, M., Schumacher, D., Stark, K., Höhle, M. (2015), in revision.

Examples

data(salmAllOnset)
rangeTest <- 410:412
alpha <- 0.05
# Control slot for the proposed algorithm with D=0 correction
controlNormal <- list(range = rangeTest, b = 4, w = 3,
                      pastAberrations = TRUE, mc.munu=10, mc.y=10,
                      verbose = FALSE,
                      alpha = alpha, trend = TRUE,
                      limit54=c(0,50),
                      noPeriods = 10, pastWeeksNotIncluded = 26,
                      delay=FALSE,inferenceMethod="asym")
# Control slot for the proposed algorithm with D=10 correction
controlDelay <-  list(range = rangeTest, b = 4, w = 3,
                      pastAberrations = TRUE, mc.munu=10, mc.y=10,
                      verbose = FALSE,
                      alpha = alpha, trend = TRUE,
                      limit54=c(0,50),
                      noPeriods = 10, pastWeeksNotIncluded = 26,
                      delay=TRUE,inferenceMethod="asym")
salm.Normal <- bodaDelay(salmAllOnset, controlNormal)
salm.delay <- bodaDelay(salmAllOnset, controlDelay)
plot(salm.Normal)
plot(salm.delay)

jimhester/surveillance documentation built on May 19, 2019, 10:33 a.m.