Truncation Distribution Options

Returns a truncation distribution formatted for usage by downstream functions. See estimate_truncation() for an approach to estimate these distributions.

Usage

trunc_opts(dist = Fixed(0), default_cdf_cutoff = 0.001, weight_prior = FALSE)

Arguments

dist

A delay distribution or series of delay distributions reflecting the truncation. It can be specified using the probability distributions interface in EpiNow2 (See ?EpiNow2::Distributions) or estimated using estimate_truncation(), which returns a dist object, suited for use here out-of-box. Default is a fixed distribution with maximum 0, i.e. no truncation.

default_cdf_cutoff

Numeric; default CDF cutoff to be used if an unconstrained distribution is passed as dist. If dist is already constrained by having a maximum or CDF cutoff this is ignored.

weight_prior

Logical; if TRUE, the truncation prior will be weighted by the number of observation data points, in doing so approximately placing an independent prior at each time step and usually preventing the posteriors from shifting. If FALSE (default), no weight will be applied, i.e. the truncation distribution will be treated as a single parameter.

Value

A <trunc_opts> object summarising the input truncation distribution.

See also

convert_to_logmean() convert_to_logsd() bootstrapped_dist_fit() Distributions

Examples

# no truncation
trunc_opts()
#> - fixed value:
#>   0

# truncation dist
trunc_opts(dist = LogNormal(mean = 3, sd = 2, max = 10))
#> - lognormal distribution (max: 10):
#>   meanlog:
#>     0.91
#>   sdlog:
#>     0.61