BayesOutbreak: the Bayes outbreak detector of
surveillance::algo.bayes. CusumSurveillance: CUSUM of
standardised counts (algo.cusum); with expected counts per region
the spatial surveillance CUSUM. EcologicalRegression: Poisson,
negative binomial and zero-inflated Poisson regression with expected-count
offsets. LerouxPrecision and Bym2Structure: Leroux and BYM2
area-effect structures. BufferExposure and KernelExposure:
exposure assessment. Identical to the Python arm morie.fn.episurv.
Usage
BayesOutbreak(
observed,
freq = 52,
b = 0,
w = 6,
act_y = TRUE,
alpha = 0.05,
time_points = NULL
)
CusumSurveillance(
observed,
expected = NULL,
start = NULL,
k = 1.04,
h = 2.26,
trans = "standard",
reset = FALSE
)
EcologicalRegression(
counts,
expected,
X = NULL,
family = "poisson",
Z = NULL,
tol = 1e-12,
maxit = 500
)
LerouxPrecision(A, rho, tau = 1)
Bym2Structure(A, tau = 1, phi = 0.5, tol = 1e-10)
BufferExposure(
receptors,
sources,
radius,
weights = NULL,
metric = "euclidean"
)
KernelExposure(
receptors,
sources,
bandwidth,
weights = NULL,
kernel = "gaussian"
)Arguments
- observed
Counts (vector, or times by regions matrix).
- freq
Observations per year.
- b
Number of past years.
- w
Half-window width.
- act_y
Use the preceding w counts.
- alpha
Test level.
- time_points
1-based time points to evaluate.
- expected
Expected counts (same shape as observed).
- start
First monitored time point when no expected counts are given.
- k, h
CUSUM reference value and threshold.
- trans
"standard","rossi","anscombe"or"none".- reset
Reset the sum after an alarm.
- counts
Area counts.
- X
Area covariates (no intercept).
- family
"poisson","negbin"or"zip".- Z
Zero-inflation covariates (no intercept).
- tol, maxit
Convergence controls.
- A
Symmetric adjacency matrix.
- rho
Leroux spatial parameter.
- tau
Precision.
- phi
Share of structured variance.
- receptors, sources
Coordinates (rows).
- radius
Buffer radius.
- weights
Source weights.
- metric
"euclidean"or"haversine".- bandwidth
Kernel bandwidth.
- kernel
"gaussian"or"quartic".
References
Hoehle, M. (2007). surveillance: an R package for the monitoring of infectious diseases. Computational Statistics 22, 571-582.
Rogerson, P. A. and Yamada, I. (2004). Approaches to syndromic surveillance when data consist of small regional counts. MMWR 53, 79-85.
Lambert, D. (1992). Zero-inflated Poisson regression. Technometrics 34, 1-14.
Leroux, B. G., Lei, X. and Breslow, N. (2000). Estimation of disease rates in small areas: a new mixed model for spatial dependence. Springer, 179-191.
Riebler, A., Sorbye, S. H., Simpson, D. and Rue, H. (2016). An intuitive Bayesian spatial model for disease mapping that accounts for scaling. Statistical Methods in Medical Research 25, 1145-1165.