Taha-Rouabah
commented
3 years ago Salut Boud,
Merci pour cette notification. Ces informations pourront nous être utile pour un éventuel second papier sur le sujet. Content que ton papier soit en phase finale de publication et que le processus t”a permis de l'améliorer. Tiens moi informé lorsque version finale sera publiée ;)
Amicalement. Taha
On May 7, 2021, at 12:48 AM, broukema @.***> wrote:
In case anyone is interested in the credibility of the national-level daily confirmed cases, see the Appendix of ArXiv:2007.11779 https://arxiv.org/abs/2007.11779 = zenodo.4432080 https://zenodo.org/record/4432080 (nearly ready for the next round in the peer review process). This shows that the JHU CSSE daily count data give similar results to the Wikipedia COVID-19 Case Count Task Force (C19CCTF) daily count data: a small bunch of countries appear to have suspiciously low statistical noise. There are some minor differences in the results - e.g. in commit 877a917, Saudi Arabia is detected as having some of the most suspiciously low-noise data in the JHU CSSE data, in Jan 2021 (7-day sequence) and Feb 2021 (14-day sequence), but is selected in 2020 for a suspiciously low-noise period in the C19CCTF analysis.
Anyone wishing to double-check that they get the same results (this aims to be a reproducible paper, so independent people should get the same results on any unix-like system) before I submit the revised version that responds to the reviewers' comments is welcome to check the full package and post issues https://codeberg.org/boud/subpoisson/issues at the git repository: https://codeberg.org/boud/subpoisson/src/branch/postreferee_fixes https://codeberg.org/boud/subpoisson/src/branch/postreferee_fixes . The "verification" step is currently turned off, but you should be able to produce the pdf file with the results, starting from the source package.
@Taha-Rouabah https://github.com/Taha-Rouabah Salut ! :) You may be interested in the update - late August 2020+September 2020 are best modelled as subpoissonian for Algeria depending on whether the C19CCTF or JHU CSSE data are preferred. My reviewers' comments have forced me to do some more work, but they've helped to improve the paper and somewhat strengthen the results. If the paper is accepted, then I'll opt for the full reviewing pipeline to become public (this is encouraged by the journal).
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/CSSEGISandData/COVID-19/issues/4057, or unsubscribe https://github.com/notifications/unsubscribe-auth/APM3UG25PCNY7JGBTYYGE2LTMMTFJANCNFSM44IE6GIQ.
broukema
In case anyone is interested in the credibility of the national-level daily confirmed cases, see the Appendix of ArXiv:2007.11779 = zenodo.4432080 (nearly ready for the next round in the peer review process). This shows that the JHU CSSE daily count data give similar results to the Wikipedia COVID-19 Case Count Task Force (C19CCTF) daily count data: a small bunch of countries appear to have suspiciously low statistical noise. There are some minor differences in the results - e.g. in commit 877a917, Saudi Arabia is detected as having some of the most suspiciously low-noise data in the JHU CSSE data, in Jan 2021 (7-day sequence) and Feb 2021 (14-day sequence), but is selected in 2020 for a suspiciously low-noise period in the C19CCTF analysis.
Anyone wishing to double-check that they get the same results (this aims to be a reproducible paper, so independent people should get the same results on any unix-like system) before I submit the revised version that responds to the reviewers' comments is welcome to check the full package and post issues at the git repository: https://codeberg.org/boud/subpoisson/src/branch/postreferee_fixes . The "verification" step is currently turned off, but you should be able to produce the pdf file with the results, starting from the source package.
@Taha-Rouabah Salut ! :) You may be interested in the update - late August 2020+September 2020 are best modelled as subpoissonian for Algeria depending on whether the C19CCTF or JHU CSSE data are preferred. My reviewers' comments have forced me to do some more work, but they've helped to improve the paper and somewhat strengthen the results. If the paper is accepted, then I'll opt for the full reviewing pipeline to become public (this is encouraged by the journal). Super-poissonian and Poissonian models for the Algerian data are significantly rejected. See commit 81da1aef12 in _postrefereefixes .