The SIR model fitted to COVID-19 infected cases data from Wuhan. Apply parameters of SIR to UK for prediction.
`wuhan_SIR_model.py' fits parameters in SIR model: gamma and beta
- gamma: contact parameter = 1/R0
- beta: recovery rate = 1/days-required-to-recover
Current fitted parameter:
| gamma | beta |
|---|---|
| 0.33838125 | 0.06476182 |
| R0 | recovery-day |
|---|---|
| 2.9552465186097683 | 15.441196085990235 |
Covariance matrix:
| pcov | gamma | beta |
|---|---|---|
| gamma | 5.10117057e-06 | 4.55526000e-07 |
| beta | 4.55526000e-07 | 9.14512396e-07 |
Last update: 15 Mar 2020
Assume suspected population of south korea is 51/11 larger than Wuhan. Data with selected days fitted to SIR model. The above figure showed the fast outbreak part of the confirmed cases between 30th Jan and 24th Feb (25 days).
Fitted parameters:
| gamma | beta |
|---|---|
| 1.43634329 | 1.18480972 |
| R0 | recovery-day |
|---|---|
| 0.6962123914245263 | 0.8440173825556665 |
Note: R0 in S Korea seemed to be smaller than 1.
Covariance matrix:
| pcov | gamma | beta |
|---|---|---|
| gamma | 5.80282701e-05 | 5.65953981e-05 |
| beta | 5.65953981e-05 | 5.60933292e-05 |
Last update: 23 Mar 2020
Note: y axis is log10 of infected cases.
Last update: 23 Mar 2020
It seemed that R0 in Italy is roughly higher than 3 (Wuhan model).
Last update: 23 Mar 2020