# COVID-19: Europe

https://wp.me/p7ciWq-mi

Analysing the incidence of population density on COVID19 infection in Europe

Multiple Regression – COVID19 Confir./1000 inh. 04/24
Dependent variable: COVID19 Confir./1000 inh. 04/24
Independent variables:
Population
density (per km2)
Area (km2)

```.                                 Standard       T
Parameter                Estimate   Error    Statistic   P-Value
CONSTANT                 0.790899  0.39803    1.98703    0.0580
density (per km2)        0.0057684 0.00246611 2.33907    0.0276

Analysis of Variance
Source                Sum of Squares Df   Mean Square F-Ratio P-Value
Model                   8.10235      1     8.10235     5.47   0.0276
Residual               37.0223      25     1.48089
Total (Corr.)          45.1246      26

R-squared = 17.9555 percent
R-squared (adjusted for d.f.) = 14.6737 percent
Standard Error of Est. = 1.21692
Mean absolute error = 0.928434
Durbin-Watson statistic = 1.77972 (P=0.2894)
Lag 1 residual autocorrelation = 0.108316```

Stepwise regression
Method: forward selection
P-to-enter: 0.05
P-to-remove: 0.05

# COVID-19 : Hospital Bed Requirements and Testing Strategy

https://wp.me/p7ciWq-lZ

## Source

This analysis is based on France data : COVID-19 Cases in hospitals 14/04/2020, by “département”.

This is an analysis based on public data, and subject to revisions or errors including the processing.

Data sources: Géodes – données en Santé Publique, INSEE.

#### Warning

As a consequence of patients being displaced to hospitals in other départements than the origin,  the number of beds decreased artificially in places (in the “Est région” mainly) and artificially increased in others (e.g. Loire ?).

What is the utility of publishing area (“départements”) statistics, if they do not reflect the cases originated in the area ?

This has an impact mainly on the relationship between population density and hospitalisation requirements. It however seems negligible in the present analysis.

## 1. Hospital bed requirements

This analysis stresses the major role plaid by population density as a factor of COVID-19 transmission. It leads to a distinction between 4 areas, according to population density, with practical consequences on the demand of equipped beds. I refer to chart 1

Numbers of beds are rough estimates based on the slope of the regression line.

1 – Areas with a population density lower than 1000 inhabitants per km² (0.386 square mile) : COVID-19 bed requirements : none

2 – Areas with a population density between 1000 and 4000 inhabitants per km² (0.386 square mile) : COVID-19 bed requirements : 0.5 per 1000 inhabitants.

3 – Areas with a population density between 4000 and 12000 inhabitants per km² (0.386 square mile) : COVID-19 bed requirements : 1.5 per 1000 inhabitants.

4 – Areas with a population density above 12000 inhabitants per km² (0.386 square mile) : COVID-19 bed requirements : 2.5 per 1000 inhabitants. In this category are cities such as Paris, New York, Wuhan, etc.

Of course, due to the fact that the outbreak was sudden – at least felt as sudden – it led to an overwhelming of the health system in the largest cities of the world. But measures designed to slow down the circulation of people in these areas – and in and out of these areas -, including total lockdowns, have been effective and led to dramatically reduce the transmission rate of the virus, hence hospitalisation requirements However, because of the average length of hospitalisation being above 10 days, this doesn’t immediately show, neither in hospitalisation-, nor, in deaths- statistics.

## 2. Testing Strategy

It seems impossible in most countries to test a large number of people over a short span of time. This would however provide the appropriate knowledge of the degree of infection of the entire population.

Alternatively, sampling techniques can be used. These are techniques used in political polls. The main difficulty of such a methodology is that, to be able to generate valid statistical inferences, it requires a random choice of the sample over the entire population, which is not cost efficient. Therefore, most such surveys are in effect run on basis of reasoned choices, i.e. quotas or ratios defining clusters drawn from the entire population which in an aggregated form are assumed to be valid representations of the entire population, as would be, for example in politics, electoral constituencies representing the national choice.

Concerning COVID-19 testing, I propose to choose sampling clusters according to the density of population criterium, distinguishing the afore mentioned 4 categories of density, each with 5000 tests randomly chosen in the category. Of course, the estimate would gain in accuracy if the number of tests is increased in each cluster. And this would be facilitated in areas where population is concentrated. Hence, a variation of the methodology could be :

10000 tests randomly chosen in areas of category 4

3000 tests randomly chosen in areas of category 3

2000 tests randomly  chosen  in areas of category 2

1000 tests randomly chosen in areas of category 1

## 3. Analysis

Both population and density of population are significant factors in explaining the COVID-29 outbreak in France. The adjusted R squared, now (i.e. patients in hospitals to April 14th 2020) is slightly above 80%,  close to reach the norm of 85%  (see also F and T statistics).

The slopes of the least squares adjusted lines continue rising, especially concerning population density. The concentration in population is the main influence behind the increases. The coefficient has risen, so far, to 0,13. It is to be stressed that it represents accumulated hospital entries (net of outgoings) between March 18 and April 14. This reflects the length of the average hospitalisation time (more than 10 days). It represents the increase in hospitals bed requirements, when population density increases by 1%. In other words, for every 10%  increase in population density, there is a need for 1.3 additional hospitalisation bed resulting from COVID-19 infection in highly populated areas (above our previous estimate of 1). E.g. with a density of 20860, the needs for Paris are of some additional 2712 beds, as of April 14th .

The incidence of population remains second, and has only slightly increased, from a coefficient of 0.3 per thousand in the previous estimate to 0,4.

### 3.2. Statistical modelling

`Multiple Regression - COVID19 in hospitals 200414 Dependent variable: COVID19 in hospitals 200414 Independent variables: Density Population .    .                             Standard           T  Parameter      Estimate          Error         Statistic      P-Value CONSTANT      -61.9653         38.9732        -1.58995        0.1151 Density         0.128287        0.0108429     11.8314         0.0000 Population      0.000468948     0.0000512331   9.15322        0.0000 Analysis of Variance Source         Sum of Squares     Df      Mean Square  F-Ratio    P-Value Model          2.15985E7          2       1.07993E7     203.49     0.0000  Residual       5.2008E6          98        53069.3 Total (Corr.)  2.67993E7        100  R-squared = 80.5936 percent R-squared (adjusted for d.f.) = 80.1975 percent Standard Error of Est. = 230.368 Mean absolute error = 154.588 Durbin-Watson statistic = 1.46077 (P=0.0031) Lag 1 residual autocorrelation = 0.26664`

The output shows the results of fitting a multiple linear regression model to describe the relationship between COVID19 in hospitals 200414 and 2 independent variables.

The equation of the fitted model is : ‘ f ne ?” XD en ces Cède à

COVID19 in hospitals 200414 = -61.9653 + 0.128287*Density + 0.000468948*Population

Since the P-value in the ANOVA table is less than 0.05, there is a statistically significant relationship between the variables at the 95.0% confidence level. The R-Squared statistic indicates that the model as fitted explains 80.5936% of the variability in COVID19 in hospitals 200414. The adjusted R-squared statistic, which is more suitable for comparing models with different numbers of independent variables, is 80.1975%. The standard error of the estimate shows the standard deviation of the residuals to be 230.368. This value can be used to construct prediction limits for new observations by selecting the Reports option from the text menu. The mean absolute error (MAE) of 154.588 is the average value of the residuals. The Durbin-Watson (DW) statistic tests the residuals to determine if there is any significant correlation based on the order in which they occur in your data file. Since the P-value is less than 0.05, there is an indication of possible serial correlation at the 95.0% confidence level.

NB. The serial autocorrelation is mainly explained

model can be simplified, notice that the highest P-value on the independent variables is 0.0000, belonging to Population. Since the P-value is less than 0.05, that term is statistically significant at the 95.0% confidence level.

### 3.3. Charts

With the extension of the infection, density of population clearly appears to be the dominant factor. A lesson to be learned for future urbanistic requirements.

#### Chart 1

The above chart is the most interesting. It clearly shows the major incidence of population density, After the exceptional cases in the East of France départements (here appearing with a diminishing role in the total picture), The high density of the Paris area and other nearby départements (Hauts-de-Seine, Seine-Saint-Denis, Val-de-Marne…) are driving forces and, just after, Bouches-du-Rhône (Marseille is the second populated city in France) and Rhône (Lyon, third populated city).

# COVID-19 : Hospital Bed Requirements and Testing Strategy

https://wp.me/p7ciWq-lC

## Source

This analysis is based on France data : COVID-19 Cases in hospitals 14/04/2020, by “département”.

This is an analysis based on public data, and subject to revisions or errors including the processing.

Data sources: Géodes – données en Santé Publique, INSEE.

#### Warning

As a consequence of patients being displaced to hospitals in other départements than the origin,  the number of beds decreased artificially in places (in the “Est région” mainly) and artificially increased in others (e.g. Loire ?).

What is the utility of publishing area (“départements”) statistics, if they do not reflect the cases originated in the area ?

This has an impact mainly on the relationship between population density and hospitalisation requirements. It however seems negligible in the present analysis.

## 1. Hospital bed requirements

This analysis stresses the major role plaid by population density as a factor of COVID-19 transmission.It leads to a distinction between 4 areas, according to population density, with practical consequences on the demand of equipped beds . We refer to chart 1

Numbers of beds are rough estimates based on the slope of the regression line.

1 – Areas with a population density lower than 1000 inhabitants per km² (0.386 square mile) : COVID-19 bed requirements : none

2 – Areas with a population density between 1000 and 4000 inhabitants per km² (0.386 square mile) : COVID-19 bed requirements : 0.5 per 1000 inhabitants.

3 – Areas with a population density between 4000 and 12000 inhabitants per km² (0.386 square mile) : COVID-19 bed requirements : 1.5 per 1000 inhabitants.

4 – Areas with a population density above 12000 inhabitants per km² (0.386 square mile) : COVID-19 bed requirements : 2.5 per 1000 inhabitants. In this category are cities such as Paris, New York, Wuhan, etc.

Of course, due to the fact that the outbreak was sudden – at least felt as sudden – it led to an overwhelming of the health system in the largest cities of the world. But measures designed to slow down the circulation of people in these areas – and in and out of these areas -, including total lockdowns have been effective and led to dramatically reduce the transmission rate of the virus, hence hospitalisation requirements However, because of the average length of hospitalisation being above 10 days, this doesn’t immediately show, neither in hospitalisation, nor, in deaths- statistics.

## 2. Testing Strategy

It seems impossible in most countries to test a large number of people over a short span of time. This would however provide the appropriate knowledge of the degree of infection of the entire population.

Alternatively, sampling techniques can be used. These are techniques used in political polls. The main difficulty of such a methodology is that, to be able to generate valid statistical inferences, it requires a random choice of the sample over the entire population, which is not cost efficient. Therefore, most such surveys are in effect run on basis of reasoned choices, i.e. quotas or ratio defining clusters on the entire population which in an aggregated form are assumed to be valid representations of the entire population, as would be, for example in politics, electoral constituencies representing the national choice.

Concerning COVID-19 testing, what I propose, here, is to choose sampling clusters according to the density of population criterium, distinguishing the afore mentioned 4 categories of density, each with 5000 tests randomly chosen in the category. Ohere thef course, the estimate would gain in accuracy if the number of tests is increased in each clusters. And this would be facilitated in areas where population is concentrated. Hence, a variation of the methodology could be

10000 tests randomly chosen in areas of category 4

3000 tests randomly chosen in areas of category 3

2000 tests randomly  chosen  in areas of category 2

1000 tests randomly chosen in areas of category 1

## 3. Analysis

Both population and density of population are significant factors in explaining the COVID-29 outbreak in France. The adjusted R squared, now (i.e. patient in hospitals to April 14th 2020) is slightly above 80%,  close to reach the norm of 85%  (see also F and T statistics).

The slopes of the least squares adjusted lines continue rising, especially concerning population density. The concentration in population is the main influence behind the increases. The coefficient has risen, so far, to 0,13. It is to be stressed that it represents accumulated hospital entries (net of outgoings) between March 18 and April 14. This reflects the length of the average hospitalisation time (more than 10 days). It represents the increase in hospitals bed requirements, when population density increases by 1%. In other words, for every 10%  increase in population density, there is a need for 1.3 additional hospitalisation bed resulting from COVID-19 infection in highly populated areas (above our previous estimate of 1). E.g. with a density of 20860, the needs for Paris are of some additional 2712 beds, as of April 14th .

The incidence of population remains second, and has only slightly increased, from a coefficient of 0.3 per thousand in the previous estimate to 0,4.

### 3.2. Statistical modelling

```Multiple Regression - COVID19 in hospitals 200414
Dependent variable: COVID19 in hospitals 200414
Independent variables:
Density
Population

.                                   Standard        T
Parameter           Estimate          Error       Statistic       P-Value
CONSTANT           -61.9653         38.9732        -1.58995        0.1151
Density              0.128287        0.0108429     11.8314         0.0000
Population           0.000468948     0.0000512331   9.15322        0.0000

Analysis of Variance
Source      Sum of Squares      Df        Mean Square      F-Ratio    P-Value
Model           2.15985E7        2         1.07993E7       203.49     0.0000
Residual        5.2008E6        98          53069.3
Total (Corr.)   2.67993E7      100

R-squared = 80.5936 percent
R-squared (adjusted for d.f.) = 80.1975 percent
Standard Error of Est. = 230.368
Mean absolute error = 154.588
Durbin-Watson statistic = 1.46077 (P=0.0031)
Lag 1 residual autocorrelation = 0.26664

The output shows the results of fitting a multiple linear regression model
to describe the relationship between COVID19 in hospitals 200414 and 2
independent variables. The equation of the fitted model is
COVID19 in hospitals 200414 =
-61.9653 + 0.128287*Density + 0.000468948*Population

Since the P-value in the ANOVA table is less than 0.05, there is a
statistically significant relationship between the variables at
the 95.0% confidence level.

The R-Squared statistic indicates that the model as fitted explains
80.5936% of the variability in COVID19 in hospitals 200414.
The adjusted R-squared statistic, which is more suitable for comparing
models with different numbers of independent variables, is 80.1975%.
The standard error of the estimate shows the standard deviation of the
residuals to be 230.368. This value can be used to construct prediction
imits for new observations by selecting the Reports option from the text
menu. The mean absolute error (MAE) of 154.588 is the average value of
the residuals. The Durbin-Watson (DW) statistic tests the residuals to
determine if there is any significant correlation based on the order
in which they occur in your data file. Since the P-value is less
than 0.05, there is an indication of possible serial correlation at
the 95.0% confidence level.

NB. The serial autocorrelation is mainly explained by the outbreak in
the "Région Est départements" (see Chart 4)
In determining whether the model can be simplified, notice that the
highest P-value on the independent variables is 0.0000, belonging to
Population. Since the P-value is less than 0.05, that term is
statistically significant at the 95.0% confidence level.
```

### 3.3. Charts

With the extension of the infection, density of population clearly appears to be the dominant factor. A lesson to be learned for future urbanistic requirements.

#### Chart 1

The above chart is the most interesting. It clearly shows the major incidence of population density, After the exceptional cases in the East of France départements (here appearing with a diminishing role in the total picture), The high density of the Paris area and other nearby départements (Hauts-de-Seine, Seine-Saint-Denis, Val-de-Marne…) are driving forces and, just after, Bouches-du-Rhône (Marseille is the second populated city in France) and Rhône (Lyon, third populated city).

# France : covid-19 07/04/2020

https://wp.me/p7ciWq-ls

Two charts to illustrate where we stand in France on the 7th of April 2020

They are based on public data, and subject to revisions or errors including the processing.

Data sources: Géodes – données en Santé Publique

# France : COVID19, Cases in hospitals 03/04/2020

https://wp.me/p7ciWq-l9

## Warning

It seems to me, as a consequence of patients (about 500 at the date this is published) being displaced to hospitals in other départements than the origin,  that the number of beds is artificially decreasing in places (in the “Est région” mainly) and artificially increasing in others (e.g. Loire ?).

What is the utility of publishing area (department) statistics, if they do not reflect the cases originated in the area ?

Both population and density of population are significant factors in explaining the COVID-29 outbreak in France. With an adjusted R squared of  79.5%, the regression is close to reach the norm of 85%  (see also F and T statistics)

The slopes of the least squares adjusted lines continue rising, especially concerning population density. The concentration in population is the main influence behind the expected increases. The coefficient has risen, sofar, to 0,11. This represents the incremental increase in hospitals bed requirements, when population density increases by 1%. In other words, for every 10%  increase in population density, there is a need of 1 additional hospitalisation bed resulting from COVID19 infection. E.g. with a density of 20860, the needs for Paris are of some additional 2000 beds, as of April 3rd .

This is an analysis based on public data, and subject to revisions or errors including the processing.

Data sources: Géodes – données en Santé Publique, INSEE.

## Analysis

Multiple Regression – COVID19 in hospitals 200403
Dependent variable: COVID19 in hospitals 200403
Independent variables:
Density
Population

.                                                                            Standard                    T
Parameter                                    Estimate           Error                Statistic            P-Value
CONSTANT                                   -51.2041          34.057                  -1.50348           0.1359
Density                                            0.111556         0.00947515      11.7735             0.0000
Population                                      0.000394141  0.0000447703    8.80362           0.0000

Analysis of Variance
Source                                      Sum of Squares   Df      Mean Square     F-Ratio  P-Value
Model                                         1.58856E7             2           7.94278E6          196.00   0.0000
Residual                                    3.97145E6           98            40525.0
Total (Corr.) 1.9857E7                                       100

R-squared = 79.9998 percent
R-squared (adjusted for d.f.) = 79.5916 percent
Standard Error of Est. = 201.308
Mean absolute error = 131.79
Durbin-Watson statistic = 1.38195 (P=0.0008)
Lag 1 residual autocorrelation = 0.305956

The output shows the results of fitting a multiple linear regression model to describe the relationship between COVID19 in hospitals 200403 and 2 independent variables. The equation of the fitted model is

COVID19 in hospitals 200403 = -51.2041 + 0.111556*Density + 0.000394141*Population

Since the P-value in the ANOVA table is less than 0.05, there is a statistically significant relationship between the variables at the 95.0% confidence level.

The R-Squared statistic indicates that the model as fitted explains 79.9998% of the variability in COVID19 in hospitals 200403. The adjusted R-squared statistic, which is more suitable for comparing models with different numbers of independent variables, is 79.5916%. The standard error of the estimate shows the standard deviation of the residuals to be 201.308. This value can be used to construct prediction limits for new observations by selecting the Reports option from the text menu. The mean absolute error (MAE) of 131.79 is the average value of the residuals. The Durbin-Watson (DW) statistic tests the residuals to determine if there is any significant correlation based on the order in which they occur in your data file. Since the P-value is less than 0.05, there is an indication of possible serial correlation at the 95.0% confidence level. Plot the residuals versus row order to see if there is any pattern that can be seen.

In determining whether the model can be simplified, notice that the highest P-value on the independent variables is 0.0000, belonging to Population. Since the P-value is less than 0.05, that term is statistically significant at the 95.0% confidence level. Consequently, you probably don’t want to remove any variables from the model.

# Italy : COVID-19, cases to 30/03/2020

https://wp.me/p7ciWq-kV

## General comment

The momentum data, from 16/03 to 30/03, confirms the dramatic role of the most populous regions of Italy, i.e. Lombardia in the first place and Emilia-Romagna and Piemonte in the second and third, but much lesser extend. Other regions behave quasi similarly , from one to the other. A regression only on them would give a line with a rather low slope.

Notice that the slope of the regression line (on all regions) has increased significantly between the 16th and the 30th of March. The peak of the pandemic in Italy is to be fixed at around the mid of March (see Figura 1 and 2 of the report quoted below in the sources).

This is an analysis based on public data, and subject to revisions or errors including the processing.

Data sources: Epidemia COVID-19 30 marzo 2020 – ore 16:00 – Istituto  Superiore di Sanita (ISS), Roma, and regional data concerning demographics and areas

## Analysis I – Cumulative cases to 30/03/2020

### Analysis

Multiple Regression – COVID-19 Cases 30/03/20
Dependent variable: COVID-19 Cases 30/03/20
Independent variables:
Population (01-2019)
hab/km²

.                                                                    Standard             T
Parameter                       Estimate                Error         Statistic           P-Value
CONSTANT                       -3314.87               472.75         -1.34056         0.1967
Population (01-2019 ) 0.00254471     0.000633961        4.01398        0.0008

Analysis of Variance
Source                            Sum of Squares            Df     Mean Square F-Ratio P-Value
Model                                 7.90733E8                  1            7.90733E8    16.11    0.0008
Residual                             8.83389E8               18             4.90772E7
Total (Corr.)                      1.67412E9                19

R-squared = 47.2327 percent
R-squared (adjusted for d.f.) = 44.3012 percent
Standard Error of Est. = 7005.51
Mean absolute error = 4900.97
Durbin-Watson statistic = 1.8902 (P=0.4265)
Lag 1 residual autocorrelation = 0.012704

Stepwise regression
Method: forward selection
P-to-enter: 0.05
P-to-remove: 0.05

## Analysis II – Cases from 16/03/2020 to 30/03/2020

### Analysis

Multiple Regression – COVID-19 Cases 30/03/20-COVID-19 Cases 16/3/2020
Dependent variable: COVID-19 Cases 30/03/20-COVID-19 Cases 16/3/2020
Independent variables:
Population (01-2019)
km²

.                                                                                Standard                T
Parameter                                  Estimate                 Error            Statistic          P-Value
CONSTANT                                – 1817.49                 1708.3           -1.06391          0.3014
Population (01-2019)           0.00163341           0.000437973       3.72947          0.0015

Analysis of Variance
Source                               Sum of Squares         Df            Mean Square   F-Ratio     P-Value
Model                                    3.25794E8                1                 3.25794E8         13.91     0.0015
Residual                                  4.2162E8             18                  2.34233E7
Total (Corr.) 7.47414E8 19

R-squared = 43.5895 percent
R-squared (adjusted for d.f.) = 40.4556 percent
Standard Error of Est. = 4839.77
Mean absolute error = 3436.17
Durbin-Watson statistic = 1.88585 (P=0.4226)
Lag 1 residual autocorrelation = -0.0204147

Stepwise regression
Method: forward selection

# France : COVID19 – Cases in Hospitals 31/03/2020

https://wp.me/p7ciWq-kx

## I – Increase in hospital bed requirements linked to COVID18, from 03/25 to 03/31

This analysis confirms the rough evaluation from the cumulative data (see part II of this post). Density appears not to be significant at this momentum. It has nevertheless been kept in the model.

This is an analysis based on public data, and subject to revisions or errors including the processing.

Data sources: Géodes – données en Santé Publique, INSEE.

### Analysis

Multiple Regression – COVID19 in hospitals 200331-COVID19 In hospitals 200325
Dependent variable: COVID19 in hospitals 200331-COVID19 In hospitals 200325
Independent variables:
Density
Inhabitants per km²

.                                                                         Standard            T
Parameter                                 Estimate           Error       Statistic    P-Value
CONSTANT                                 73.014            10.4148       7.01062   0.0000
Density                                    -0.168036          0.425494   -0.394921   0.6938
Inhabitants per km²              0.226678          0.425421      0.532832 0.5954

Analysis of Variance
Source                                  Sum of Squares      Df      Mean Square    F-Ratio     P-Value
Model                                     1.96388E6               2            981941.             95.05     0.0000
Residual                                 1.01239E6            98             10330.5
Total (Corr.)                           2.97627E6          100

R-squared = 65.9846 percent
R-squared (adjusted for d.f.) = 65.2904 percent
Standard Error of Est. = 101.639
Mean absolute error = 70.5319
Durbin-Watson statistic = 1.69812 (P=0.0650)
Lag 1 residual autocorrelation = 0.143963

## II – Analysis on cumulative data to 03/31

Both population and density of population are significant factors in explaining the COVID-29 outbreak in France. With an adjusted R squared of 76.7% , to be compared to 74.4% two days ago the regression continues to gain in signification (see also F and T statistics).  Indeed, the role of the Est Région continues to be reduced in the model, meaning that the entire French territory is following a trajectory closer to that of the Est.

The slopes of the least squares adjusted lines continue rising, especially concerning population density, from 0.076 on the 29th of March (cumulative data) to 0.084 on the 31st. In other words the concentration of population is the major influence behind the expected increases. From this it is easy to understand that the main needs in hospitalisations will be concentrated in the largest cities : Paris, Marseilles, Lyon, Nice…  The needs for Paris (and Ile-de-France) for example will exceed 7500 beds with ventilators in the coming days.

This is an analysis based on public data, and subject to revisions or errors including the processing.

Data sources: Géodes – données en Santé Publique, INSEE.

### Analysis

Multiple Regression – COVID19 in hospitals 200331
Dependent variable: COVID19 in hospitals 200331
Independent variables:
Inhabitants per km²
Population

.                                                                        Standard                  T
Parameter                             Estimate             Error                Statistic          P-Value
CONSTANT                            -39.4282            30.1763              -1.30659           0.1944
Inhabitants per km²             0.0898564        0.00839416      10.7046             0.0000
Population                              0.000325191    0.0000396694   8.19752           0.0000

Analysis of Variance
Source                                 Sum of Squares         Df        Mean Square          F-Ratio    P-Value
Model                                   1.05145E7                  2             5.25727E6              165.32     0.0000
Residual                              3.11647E6                 98            31800.7
Total (Corr.) 1.3631E7                                       100

R-squared = 77.1369 percent
R-squared (adjusted for d.f.) = 76.6703 percent
Standard Error of Est. = 178.328
Mean absolute error = 112.192
Durbin-Watson statistic = 1.36107 (P=0.0005)
Lag 1 residual autocorrelation = 0.315876