flumomoStan: A bayesian version of the FluMOMO algorithm

In thl-mjv/momoStan: Bayesian analysis for the European Euromomo project using Stan

Description

Fits a regression model to a mortality time series using any number of covariates with effects constrained to be positive.

Usage

flumomoStan(data, spring = 16:25, autumn = 31:47, datevar = "date",
  mortvar = "n", covar = NULL, seasvar = NULL, popvar = NA,
  byvar = NA, iter = 2000, positive = c("all", "none", "some"),
  positives = NULL, data.only = FALSE)

Arguments

data	a data frame with all necessary variables
datevar	name of the variable containing the date of the observation (preferrably the date of the Monday)
mortvar	name of the variable containing the number of deaths
covar	character vector of covariates included in the model as such
seasvar	character vector of covariates included in the model by season
popvar	(optionally) the name of the variable containing the population denominator
byvar	(optionally) the name of the variable by which groups the analysis is done
iter	number of iterations
positive	which covariates should be restricted to positives, either all, none or some
positives	if positive=="some", which. Treated as list of regular expressions, anything that matches is treated as a hit
data.only	if TRUE just collect all the data together and don't fit

Value

stan fit object

thl-mjv/momoStan documentation built on May 3, 2019, 2:57 p.m.