Skip to main content

Table 4 Regression analysis

From: The median age of a city’s residents and population density influence COVID 19 mortality growth rates: policy implications

Variables

(1)

(2)

(3)

\(\Delta \mathrm{ln}(Cum\_Deaths)\)

\(\Delta \mathrm{ln}(Cum\_Deaths)\)

\(\Delta \mathrm{ln}(Cum\_Deaths)\)

Constant

− 0.00105***

− 0.00160***

− 0.00160***

(0.00208)

(2.74 × 10–6)

(3.05 × 10–6)

MedianAge

5.93 × 10–5***

6.00 × 10–5***

6.00 × 10–5***

(6.87 × 10–7)

(4.80 × 10–7)

(5.21 × 10–7)

PopulationSize

1.39 × 10–8***

1.29 × 10–8***

1.29 × 10–8***

(< 0.01)

(< 0.01)

(< 0.01)

Population_Density × t

2.66 × 10–10***

5.65 × 10–10***

5.67 × 10–10***

(0.00160)

(< 0.01)

(< 0.01)

Dum_vaccine × Population_Density × t

− 2.20 × 10–10***

− 3.13 × 10–11

− 3.12 × 10–11

(3.40 × 10–9)

(0.408)

(0.410)

Lockdowns

0.00197***

0.00197***

(< 0.01)

(< 0.01)

Holidays

− 2.62 × 10–5

(0.769)

Observations

63,555

63,555

63,555

R-squared between estimators

0.6056

0.6583

0.6584

Calculated F-value for the regression significance

639.09***

660.45***

550.43***

Number of CityCode

152

152

152

  1. Estimation outcomes are based on the empirical model given by Eq. (2). The R-Squared between estimators gives the goodness of fit for the general equation \({\overline{y} }_{i}=\alpha +{\overline{x} }_{i}\beta +{\nu }_{i}+{\overline{\varepsilon }}_{i}\) where \({\overline{y} }_{i}={\sum }_{t}{y}_{it}/{T}_{i}\); \({\overline{x} }_{i}={\sum }_{t}{x}_{it}/{T}_{i}\); \({\overline{\epsilon }}_{i}={\sum }_{t}{\epsilon }_{it}/{T}_{i}\) (the sample mean of cities across time) and \({\nu }_{i}\) reflect generic differences across cities. p-values are given in parentheses
  2. ***p < 0.01