glmSTARMA (version 1.0.0)

chickenpox: Chickenpox Infections in Hungary

Description

Multivariate count time series consisting of weekly chickenpox infections in the districts of Hungary.

Format

chickenpox

A matrix with counts of chickenpox infections (rows = districts, columns = time points).

W_hungary

A list of matrices containing spatial weight matrices:

1. Identity matrix.

2. Row-normalized adjacency matrix of the districts.

population_hungary

A numeric matrix containing the population per 10000 inhabitants of each district over time.

Details

This dataset contains chickenpox counts in the 20 districts (NUTS 3) of Hungary over a time period of 522 weeks (from 2005 to 2014).

The row-normalized adjacency matrix indicates which districts share a common border.

The population data is only availabyle on a yearly basis and has been linearly interpolated by us to obtain weekly estimates.

The dataset is not included directly in the package. Use load_data("chickenpox") to download it.

Examples

# \donttest{
dat <- load_data("chickenpox", directory = tempdir())
chickenpox <- dat$chickenpox
population_hungary <- dat$population_hungary
W_hungary <- dat$W_hungary

covariates <- list(population = population_hungary, 
                 season_cos = SpatialConstant(cos(2 * pi / 52 * 1:522)),
                 season_sin = SpatialConstant(sin(2 * pi / 52 * 1:522)))
glmstarma(chickenpox, list(past_obs = 1), wlist = W_hungary, 
          covariates = covariates, family = vpoisson("log"))
glmstarma(chickenpox, list(past_obs = 1), wlist = W_hungary, 
          covariates = covariates, family = vnegative.binomial("log"))

mean_model <- list(past_obs = rep(1, 7))
dispersion_model <- list(past_obs = 1)
dglmstarma(chickenpox, mean_model, dispersion_model, mean_family = vquasipoisson("log"), 
           dispersion_link = "log", wlist = W_hungary, mean_covariates = covariates)
# }