Descriptive: February 2020 to 2021

Row

Hospitalizations by provice

Hospitalizations by age group

Row

Accumulated cases

Accumulated deaths

Row

Number of cases

1056639

Number of deaths

15217

% Hospitalized

2.5

Average age group in hospitals

7

Hospitalizations prediction: February 2020 to 2021

Row

Research question

Can we predict hospitalizations?

Row

Methods

Logistic Regression Model.
Predictors: Age group, Sex, Number of days in hospital, total cases reported.
Transformation to continuous dependent variable: logit.   Model training method: Leave one out cross validation.

Row

Logistic Regression model

Coefficients Std.Errors and associated p-values.
Estimate Std. Error z value Pr(>|z|)
(Intercept) -4.4646616 0.0330840 -134.949088 0
AgegroupC 0.4976298 0.0038864 128.043516 0
SexMale 0.8333376 0.0150845 55.244452 0
Number_days -0.0122331 0.0000835 -146.525503 0
Total_cases_reported 0.0002214 0.0000362 6.117184 0

Row

Predictions versus reality

0 1
FALSE 919269 20735
TRUE 3661 2943

Row

sensetivity

0.12429

specificity

0.99603

Cases forecasting

Row

Research question

Can we forecast COVID cases and deaths for the next 3 motnhs?

Row

Methods

The prophet model.

Row

Forecasting COVID cases

Forecasting COVID deaths

Row

Correlation for COVID cases model

0.99728965613571

R-squared for COVID cases model

0.994586658235282

Correlation for COVID deaths model

0.999788189674777

R-squared for COVID deaths model

0.999576424213167