Multivariate Surveillance through independent univariate algorithms
wrap.algo.Rd
This function takes an sts
object and applies an univariate
surveillance algorithm to the time series of each observational unit.
Usage
<!-- %This is the main function -->
wrap.algo(sts, algo, control,control.hook=function(k, control)
return(control),verbose=TRUE,...)
<!-- %Derived functions fixing the control object and the "algo" argument -->
bayes(sts, control = list(range = range, b = 0, w = 6, actY = TRUE,
alpha = 0.05), ...)
rki(sts, control = list(range = range, b = 2, w = 4, actY = FALSE), ...)
cusum(sts, control = list(range = range, k = 1.04, h = 2.26,
m = NULL, trans = "standard", alpha = NULL,
reset = FALSE), ...)
glrpois(sts, control = list(range = range, c.ARL = 5, S = 1, beta = NULL,
Mtilde = 1, M = -1, change = "intercept",
theta = NULL), ...)
glrnb(sts, control = list(range = range, c.ARL = 5, mu0 = NULL, alpha = 0,
Mtilde = 1, M = -1, change = "intercept",
theta = NULL, dir = c("inc", "dec"),
ret = c("cases", "value")), ...)
outbreakP(sts, control = list(range = range, k = 100,
ret = c("cases", "value"),
maxUpperboundCases = 1e5), ...)
Arguments
- sts
Object of class
sts
- algo
Character string giving the function name of the algorithm to call, e.g.
"algo.farrington"
. Calling is done usingdo.call
.- control
Control object as list. Depends on each algorithm.
- control.hook
This is a function for handling multivariate objects. This argument is a function function of integer k and the current control object and which returns the appropriate control object for region k.
- verbose
Boolean, if
TRUE
then textual information about the process is given- ...
currently ignored.
Value
An sts
object with the alarm
, upperbound
,
etc. slots filled with the results of independent and univariate
surveillance algorithm.
See also
algo.rki
, algo.farrington
,
algo.cusum
, algo.glrpois
,
algo.glrnb
, algo.outbreakP
for the exact form of the control
object.