Author: Paul Hong Lee
Title: Estimating the real-time case fatality rate of COVID-19 using Poisson mixtures model Document date: 2020_4_15
ID: g1l3gi3l_5
Snippet: In fact, a statistical modeling approach was developed 15 years ago using the SARS data [1] . The rationale behind this approach is that deaths and recovery conditional at any time point should both follow a parametric distribution (gamma distribution was used to fit the SARS CFR data). The limitation of this approach lies in the estimation of the parameters of this distribution, which require the often-non-public data of time from confirmed diag.....
Document: In fact, a statistical modeling approach was developed 15 years ago using the SARS data [1] . The rationale behind this approach is that deaths and recovery conditional at any time point should both follow a parametric distribution (gamma distribution was used to fit the SARS CFR data). The limitation of this approach lies in the estimation of the parameters of this distribution, which require the often-non-public data of time from confirmed diagnosis to death and time from confirmed diagnosis to recovery. Here, we suggest replacing the gamma distribution by Poisson distribution, as we can see below that the knowledge of time to death and time to recovery is not necessary. The proposed model requires data of deaths, recoveries, and total confirmed cases recorded in each day since the outbreak of a population. These data for COVID-19 are publicly-available [7].
Search related documents:
Co phrase search for related documents- confirm case and time knowledge: 1
- confirm diagnosis and recovery time: 1
- confirm diagnosis and time death: 1
- gamma distribution and modeling approach: 1, 2
- gamma distribution and parametric distribution: 1, 2
- gamma distribution and Poisson distribution: 1, 2, 3, 4, 5, 6
- gamma distribution and Poisson distribution gamma distribution: 1, 2
Co phrase search for related documents, hyperlinks ordered by date