4

The reproductive number $R(t)$ quantifies the expected number of secondary infections caused by an infected individual at time $t$. Thus, an R number below 1 will be likely to stop a disease like Corona.

Is there a gold standard how to calculate $R(t)$ based on daily time series data? What are the most established approaches?

I am not interested in answers like "blindly use function X in package Y" but rather in specific formulas or approaches.

kjetil b halvorsen
  • 63,378
  • 26
  • 142
  • 467
Michael M
  • 10,553
  • 5
  • 27
  • 43
  • https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/1472-6947-12-147 – assumednormal Jun 27 '20 at 16:54
  • 1
    I have seen almost a dozen way to calculate $R(t)$ (I work in a COVID-19 related project). Strictly speaking $R(t)$ is never observed and that makes it subjective; for example; we usually require at least 2 weeks of data but strictly speaking we could do it with just 1 week data. Its utility comes *after* standardisation. Gold standard: There is none. In UK if *pressed* and knowing I would be partially wrong, I would go with the [PHE/MRC](https://www.mrc-bsu.cam.ac.uk/blog/our-latest-real-time-tracking-of-covid-19/) model. Other countries may have different golden standard models.... – usεr11852 Jun 03 '21 at 11:49
  • @us%ce%b5r11852: thx for your insights. The question is almost one year old and I also gained the impression that there might not be such gold standard. – Michael M Jun 03 '21 at 11:59
  • @MichaelM: No biggie! It popped-up due to being re-tagged. :) – usεr11852 Jun 03 '21 at 22:47

0 Answers0