Effective reproduction number Reff

The effective reproduction number Reff is the average number of people infected by a single infected person.

Evolution of Reff according to different estimators

green lineReff (EpiEstim, τ=7)
blue lineReff (EpiEstim, τ=13)
Reff (LSHTM, 50%-CI)
Reff (LSHTM, 90%-CI)
The figure depicts the estimated evolution of Reff in Austria according to the software EpiEstim using two fixed values for the parameter τ. (τ=13 is used by AGES who is the main source for estimates of Reff in Austria.)
In addition, the figure displays an enhanced version of EpiEstim developed by epiforecast.io (LSHTM). For the latter estimator we depict credible intervals (50%-CI and 90%-CI). Based on case numbers reported on September 24 (last update), the value of Reff on September 14 is estimated to lie between 0.96 and 1.03 (90%-CI), with a median of 1.0.
September 14 is the most recent day for which estimates are available as Reff can only be estimated with a time lag of approximately 10 days.
Current trust in the estimator / credible interval:
The available data leaves some uncertainties that cannot be taken into account quantitatively in the estimates. In the current situation, it cannot be ruled out that these may significantly distort the results.

Case numbers

Here we show the evolution of the number of people who tested positive for SARS-CoV-2, based on data of the WHO.

Positively tested cases in Austria

Curve of positively tested cases of SARS-CoV-2 in Austria
Reporting date and infection date: In the figure above, the green bars depict the number of new positive tests (cases) reported on that day. The dark grey band gives the estimate of the positive cases assigned to the date at which the infection took place. The light grey area gives the credible interval around it.
Hypothesis test for Reff ≤ 1: In the week from September 18 to September 24, 12216 new cases were reported.
Under the assumption that Reff = 1 (and given the case numbers until September 17) we would have expected 14618 cases for this week (median).
The observed number of cases is under the assumption Reff ≤ 1.
A simple statistical test shows that under the assumption Reff = 1 (or smaller) the probablity of observing 12216 or more cases is 100% .