default_fill_missing_obs

internal

Estimates the time-varying reproduction number, rate of
spread, and doubling time using a renewal equation approach combined
with Bayesian inference via Stan. Supports Gaussian process and
random walk priors for modelling changes in
transmission over time. Accounts for delays between infection and
observation (incubation period, reporting delays), right-truncation
in recent data, day-of-week effects, and observation overdispersion.
Can estimate relationships between primary and secondary outcomes
(e.g., cases to hospitalisations or deaths) and forecast both. Runs
across multiple regions in parallel. Based on
Abbott et al. (2020) <doi:10.12688/wellcomeopenres.16006.1> and
Gostic et al. (2020) <doi:10.1101/2020.06.18.20134858>.

Sebastian Funk

EpiNow2

Estimate and Forecast Real-Time Infection Dynamics

Sam Abbott

Joel Hellewell

Katharine Sherratt

Katelyn Gostic

Joe Hickson

Hamada S. Badr

Michael DeWitt

James M. Azam

Adrian Lison

Robin Thompson

Sophie Meakin

James Munday

Nikos Bosse

Paul Mee

Peter Ellis

Pietro Monticone

Lloyd Chapman

Andrew Johnson

Kaitlyn Johnson

Adam Howes

default_fill_missing_obs function

<dl><dt>data</dt>
<dd>A <code>&lt;data.frame&gt;</code> of disease reports (confirm) by date (date).
<code>confirm</code> must be numeric and <code>date</code> must be in date format. Optionally,
<code>data</code> can also have a logical <code>accumulate</code> column which indicates whether
data should be added to the next data point. This is useful when modelling
e.g. weekly incidence data. See also the <code>fill_missing()</code> function which
helps add the <code>accumulate</code> column with the desired properties when dealing
with non-daily data. If any accumulation is done this happens after
truncation as specified by the <code>truncation</code> argument. If all entries
of <code>confirm</code> are missing (<code>NA</code>) the returned estimates will represent the
prior distributions.</dd>
<dt>obs</dt>
<dd>A list of options as generated by <code>obs_opts()</code> defining the
observation model. Defaults to <code>obs_opts()</code>.</dd>
<dt>obs_column</dt>
<dd>Character (default: "confirm"). If given, only the column
specified here will be used for checking missingness. This is useful if
using a data set that has multiple columns of hwich one of them
corresponds to observations that are to be processed here.</dd></dl>

Arguments

Temporary function to support the transition to full support of missing
data. — default_fill_missing_obs

<dl>

<dt>data</dt>
<dd>A <code>&lt;data.frame&gt;</code> of disease reports (confirm) by date (date).
<code>confirm</code> must be numeric and <code>date</code> must be in date format. Optionally,
<code>data</code> can also have a logical <code>accumulate</code> column which indicates whether
data should be added to the next data point. This is useful when modelling
e.g. weekly incidence data. See also the <code>fill_missing()</code> function which
helps add the <code>accumulate</code> column with the desired properties when dealing
with non-daily data. If any accumulation is done this happens after
truncation as specified by the <code>truncation</code> argument. If all entries
of <code>confirm</code> are missing (<code>NA</code>) the returned estimates will represent the
prior distributions.</dd>


<dt>obs</dt>
<dd>A list of options as generated by <code>obs_opts()</code> defining the
observation model. Defaults to <code>obs_opts()</code>.</dd>


<dt>obs_column</dt>
<dd>Character (default: "confirm"). If given, only the column
specified here will be used for checking missingness. This is useful if
using a data set that has multiple columns of hwich one of them
corresponds to observations that are to be processed here.</dd>

</dl>

Temporary function to support the transition to full support of missing
data.

default_fill_missing_obs: Temporary function to support the transition to full support of missing data.

Description

Usage

Value

Arguments