Table 1.
Spatial Weight Matrix Discrimination
| Weight Matrix type | Moran's I | Pseudo P-Value Moran |
|---|---|---|
| W1010km | 0.318 | 0.004 |
| W1250km | 0.209 | 0.006 |
| W2nearest | 0.699 | 0.001 |
| W4nearest | 0.536 | 0.001 |
| W5nearest | 0.487 | 0.001 |
Table 2.
Variables Description
| Variable Alias | Variable Name | Description |
|---|---|---|
| Log GDP/Cap | Development | Economic development is the endogenous variable of the study, and it's represented by a country's logged GDP/cap value |
| Transport | Public Transportation | Public transportation represents the exogenous variable of the study and was calculated as total volume of km travelled by road and rail transportation by the average citizen of the country in the year of reference |
| Greenhouse | Sustainability | Used as a control variable for sustainability, the greenhouse variable represents the total CO2 emissions per capita |
| Education | Education | A proxy variable was used to represent education, namely the graduates in tertiary education by age groups per 1000 of population between the ages of 20 and 29 |
| Attractiveness | Country Attractivity | Attractivity of the country is a dummy variable that takes the value 1 for the 5 most attractive countries in the EU in terms of investments |
| Log Roads | Infrastructure | As a proxy to represent a country's infrastructure, the logged value of the total km of roads was used |
Table 3.
Descriptive Statistics
| Variables | GDP/Cap | Transport | Sustainability | Education | Attractiveness | Log Roads |
|---|---|---|---|---|---|---|
| Mean | 4.41 | 13.13 | 7.92 | 57.24 | 0.18 | 1.97 |
| Median | 4.37 | 12.9 | 7.3 | 53.6 | 0 | 1.98 |
| Standard Error | 0.05 | 0.71 | 0.51 | 3.41 | 0.08 | 0.13 |
| Standard Deviation | 0.26 | 3.68 | 2.64 | 17.7 | 0.39 | 0.67 |
| Skewness | 0.41 | 0.34 | 1.62 | 0.56 | 1.72 | −0.61 |
| Kurtosis | −0.33 | 0.17 | 4.17 | 0.96 | 1.02 | 0.24 |
| N | 27 | 27 | 27 | 27 | 27 | 27 |
Table 4.
Correlogram
| Variables | GDP/Cap | Transport | Sustainability | Education | Attractiveness | Log Roads |
|---|---|---|---|---|---|---|
| GDP/Cap | 1 | 0.265306 | 0.678325 | 0.050464 | 0.345262 | −0.11932 |
| Transport | 0.265306 | 1 | 0.215899 | −0.44597 | 0.054979 | −0.75423 |
| Sustainability | 0.678325 | 0.215899 | 1 | −0.17053 | 0.246438 | −0.23339 |
| Education | 0.050464 | −0.44597 | −0.17053 | 1 | −0.1022 | 0.437435 |
| Attractiveness | 0.345262 | 0.054979 | 0.246438 | −0.1022 | 1 | 0.005346 |
| Log Roads | −0.11932 | −0.75423 | −0.23339 | 0.437435 | 0.005346 | 1 |
Figure 1.
Standard Deviation Map Display for Public Transport and Development
Figure 2.
Local Spatial Autocorrelation for Development
Figure 3.
Bivariate Spatial Autocorrelation for Public Transport and Development
Table 4.
Multiple OLS Regression Between Public Transport and Development
Table 5.
Diagnostics for Spatial Dependence
| Diagnostics for Spatial Dependence | Log GDP/Cap |
|---|---|
| Moran's I (errors) | 1.3554 |
| Prob | (0.17) |
| Lagrange Multiplier (lag) | 1.5050 |
| Prob | (0.21) |
| Robust LM (lag) | 6.3663 |
| Prob | (0.11) |
| Lagrange Multiplier (errors) | 0.1421 |
| Prob | (0.71) |
| Robust LM (errors) | 5.0035 |
| Prob | (0.02) |
| Lagrange Multiplier (SARMA) | 6.5084 |
| Prob | (0.03) |
Table 6.
SARMA Regression Model Between Public Transport and Development
Appendix 1.
Moran's I Pseudo P-Value for Development
Appendix 2.
Bivariate Moran's I Pseudo P-Value for Public Transport and Development