Global Journal of Human-Social Science, B: Geography, Environmental Science and Disaster Management, Volume 22 Issue 3

Road passengers (million passenger- kilometers) Passenger transport refers to the total movement of passengers using inland transport on a given network. Data are expressed in million passenger-kilometers, which represents the transport of a passenger for one kilometer. [6] . https://data.oecd.org/transport/passenger- transport. htm Railways passengers (million passenger-km) Passengers carried by railway are the number of passengers transported by rail times kilometers traveled [6] . https://data.worldbank.org/indicator/IS.RRS.PAS G. KM Urban population growth (annual %) Urban population refers to people living in urban areas as defined by national statistical offices. It is calculated using World Ban k population estimates and urban ratios fro m the United Nations World Urbanization Prospects [6]. https://data.worldbank.org/indicator/SP.URB. GROW Our study takes place over a long period (1970- 2018) and based on a large number of observations (1225 for the sample with 49 observations for each indicator and for each country). This data collection is essential to analyze the behavior of countries and measure the impact of each variable with the use of artificial intelligence and more precisely with the Panel data method. The Panel data model has a number of advantages. The double dimension of the data (individual: country and temporal: years) makes it possible to implement a monitoring algorithm which simultaneously takes into account the dynamics of behaviors and their possible heterogeneity between the countries. It constitutes an advantage over other types of method such as time series and analytic data. IV. P anel D ata The data used in artificial intelligence are most often provided by a time series. Furthermore, it is possible to have instantaneous cross-sectional data relating to a given period. Therefore, the panel data model is written as a double index ( i : individual and t : temporal) model which takes the following form Eq1: Yit = α + β Xit + ɛ it (1) • Yit is the dependent variable (CO2 emissions from transportation) • α is the intercept • β is the regression coefficient Xit is the independent variable (Population density, Agricultural land, Forest area, Road passengers, Railways passengers and Urban population) • ɛ is the error term The dual dimension offered by panel data is a major advantage. Indeed, while time series data allow us to study the evolution of relationships over time, they do not allow us to control for unobserved heterogeneity related to individuals. Conversely, cross-sectional data make it possible to analyze the heterogeneity between individuals, but they cannot take into account the dynamic behavior. Thus, by using panel data, we can exploit the two sources of variation in statistical information: Temporal where intra-individual variability (within) Individual or inter-individual variability (Between) The increase in the number of observations makes it possible to guarantee better precision of the estimators, to reduce the risks of multi-collinearity and to widen the scope of the investigation. The panel considered is not necessarily complete (balanced data) where all statistical units are observed during the period considered. This may be an incomplete, unbalanced panel where individuals are not observed over the entire period of analysis due to the input/output problem. V. R esult and D iscussion Between 1970 and 2018, global CO2 emissions from transport increased by 45% as indicated in Figure 1 and can be expected to increase by around by 70%. This trend is particularly marked in developing countries and emerging economies compared to developed countries. For example Brazil, Bolivia and Ivory Coast have more CO2 emissions due to transport compared to Australia and Canada. © 2022 Global Journals Volume XXII Issue III Version I 47 ( ) Global Journal of Human Social Science - Year 2022 B Analysis of Carbon Dioxide Emission from Transportation Sector using Panel Data Method