EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
1592. On the gap between data and models in COVID-19 analysis
Invited abstract in session WD-7: Modelling social-behavioural phenomena in creative societies, stream Behavioural OR.
Wednesday, 14:30-16:00Room: 1019 (building: 202)
Authors (first author is the speaker)
| 1. | Vytautas Dulskis
|
| Institute of Data Science and Digital Technologies, Vilnius University | |
| 2. | Leonidas Sakalauskas
|
| Vytautas Magnus University | |
| 3. | Rimas Jankunas
|
| Medical Academy, Lithuanian University of Health Sciences |
Abstract
A scrutiny analysis of the COVID-19 data is required to get insights into effective strategies for pandemic control. However, a particular gap between official data and methods used to assess the effectiveness of various containment measures (e.g., COVID-19 passports) hinders sound inference-making. Seeking to escape the burden of arising obstacles, employing the principles and methods of descriptive statistics is often tempting, but in-depth analysis demands more sensitive and reliable methods. In this regard, this paper advocates a maximum likelihood compartmental modeling approach, which provides the flexibility to raise various hypotheses about infectivity, recovery, and mortality dynamics and to find the most likely answers given the data. Our paper is based exclusively on COVID-19 deaths in light of official data limitations, as relatively fewer limitations characterize these data. Nevertheless, this paper does not solve the underlying problems but hints at potential improvements in official data reporting that could benefit COVID-19 modeling prospects.
Keywords
- Analytics and Data Science
- Dynamical Systems
- Stochastic Models
Status: accepted
Back to the list of papers