COVID-19 Data Analysis

Tasso di letalità

Tasso di letalità istantaneo ponderato (wiCFR) per COVID-19 in Italia.

Max Pierini, Alessio Pamovio, Sandra Mazzoli


Case Fatality Rate (CFR) is usually estimated as total death over total cases. This estimate though does not take into account likely dynamic variations of fatality along the time line. Is here presented a dynamic estimation of instant case fatality rate based upon onset-to-death interval: weighted instant Case Fatality Rate (wiCFR).

METODO (preprint):

Massimo Pierini, Sandra Mazzoli, & Alessio Pamovio. (2020, November 3). Estimation of weighted instant Case Fatality Rate of COVID-19 in Italy. Zenodo. http://doi.org/10.5281/zenodo.4235485


Grezzo

Filtrato

OUTLIERS in col median (0< or >0.5)
========================================
              median        Q1        Q3        lo        hi
date                                                        
2020-06-24 -0.097484 -0.147619 -0.078086 -0.190184 -0.038130
2020-08-15  0.619608  0.516340  0.818653  0.286232  1.224806

Impute function `mean`, window 7
========================================
              median        Q1        Q3        lo        hi
date                                                        
2020-06-24  0.065129  0.036507  0.093421  0.004055  0.150556
2020-08-15  0.107225  0.089046  0.141138  0.051793  0.211517

Mediana settimanale

Confronto con cumulativo

CFRcum % wiCFR %
date
2021-10-13 2.791982 1.086637
2021-10-14 2.791250 1.184483
2021-10-15 2.790525 1.268116
2021-10-16 2.789057 0.432767
2021-10-17 2.788127 0.747198
2021-10-18 2.788117 1.369863
2021-10-19 2.788008 2.258793
2021-10-20 2.786525 1.091631
2021-10-21 2.785052 1.206030
2021-10-22 2.783594 1.314016
2021-10-23 2.782123 1.327434
2021-10-24 2.780444 0.816882
2021-10-25 2.779591 1.082251
2021-10-26 2.778229 1.731602

Mediana mensile


Vaccini e wiCFR

OLS Regression Results
Dep. Variable: wiCFR R-squared: 0.564
Model: OLS Adj. R-squared: 0.562
Method: Least Squares F-statistic: 344.9
Date: Tue, 26 Oct 2021 Prob (F-statistic): 5.32e-50
Time: 17:35:19 Log-Likelihood: -171.16
No. Observations: 269 AIC: 346.3
Df Residuals: 267 BIC: 353.5
Df Model: 1
Covariance Type: nonrobust
coef std err t P>|t| [0.025 0.975]
Intercept 2.7578 0.071 38.634 0.000 2.617 2.898
due_dosi_log -0.4246 0.023 -18.572 0.000 -0.470 -0.380
Omnibus: 22.076 Durbin-Watson: 1.269
Prob(Omnibus): 0.000 Jarque-Bera (JB): 27.262
Skew: 0.622 Prob(JB): 1.20e-06
Kurtosis: 3.940 Cond. No. 8.67


Warnings:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

© 2020 Max Pierini. Thanks to Sandra Mazzoli & Alessio Pamovio. ipynb-website © 2017 Peter Carbonetto & Gao Wang

Exported from Italia/wiCFR.ipynb committed by maxdevblock on Tue Oct 26 15:35:24 2021 revision 428, cb2a8931