the limits of R
It has been repeated many times on many platforms, the R (or R⁰) number is not a great summary about the COVID-19 pandemic, see eg Rossman’s warning in The Conversation, but Nature chose to stress it one more time (in its 16 Jul edition). Or twice when considering a similar piece in Nature Physics. As Boris Johnson made it a central tool of his governmental communication policy. And some mayors started asking for their own local R numbers! It is obviously tempting to turn the messy and complex reality of this planetary crisis into a single number and even a single indicator R<1, but it is unhelpful and worse, from the epidemiology models being wrong (or at least oversimplifying) to the data being wrong (i.e., incomplete, biased and late), to the predictions being wrong (except for predicting the past). Nothing outrageous from the said Nature article, pointing out diverse degrees of uncertainty and variability and stressing the need to immediately address clusters rather than using the dummy R. As an aside, the repeated use of nowcasting instead of forecasting sounds like a perfect journalist fad, given that it does not seem to be based on a different model of infection or on a different statistical technique. (There is a nowcasting package in R, though!) And a wee bit later I have been pointed out at an extended discussion of an R estimation paper on Radford Neal’s blog.
August 11, 2020 at 5:28 pm
How confusing to write an article about the R-language with the title “The limits of R” :-) . Here I was afraid I’d find out that R couldn’t do some analysis, not that using the R that R calculates might lead to , umm, eRRoRs
August 12, 2020 at 8:37 am
the confusion in the title was sort of intentional!
August 11, 2020 at 8:12 am
[…] article was first published on R – Xi'an's Og, and kindly contributed to R-bloggers]. (You can report issue about the content on this page […]
August 10, 2020 at 8:40 pm
[…] by data_admin [This article was first published on R – Xi’an’s Og, and kindly contributed to R-bloggers]. (You can report issue about the content on this page […]
August 10, 2020 at 5:05 pm
It’s startling to me how limitations of Poisson models for modeling count data appear to be forgotten in common studies of disease dynamics, certainly in popular discussions of it. This has been recently pointed out (Hébert-Dufresne, et al, 2020, not peer reviewed, but still a good paper) and implications for SIR models drawn (Szapudi, 2020, also not peer reviewed). Worse, for the field, knowledge of these limitations is not new:
Volz EM, Miller JC, Galvani A, Ancel Meyers L (2011) Effects of Heterogeneous and Clustered Contact Patterns on Infectious Disease Dynamics. PLoS Comput Biol 7(6): e1002042.
Whether epidemiologists think the overdispersion effects are neglectable or if there is some other reason isn’t clear to me. Indications, however, are that for realistic values, policy measures based upon the Poisson model could be wrong. See
Althouse, B. M., E. A. Wenger, J. C. Miller, S. V. Scarpino, A. Allard, L. Hébert-Dufresne, and H. Hu. “Stochasticity and heterogeneity in the transmission dynamics of sars-cov-2 (2020).” arXiv preprint arXiv:2005.13689,
particularly its Figure 3.