The aim of the present diploma thesis is to investigate the impact of the pandemic on mobility in Greece, using time series analysis. On that purpose, data on mobility and restrictive measures were collected from online databases. For data analysis, ARIMA time series models were developed for Greece and Athens with dependent variable driving or walking and with exogenous factor one of the restrictive measures. The application of the methodology resulted in a significant number of models, appropriate to estimate mobility trends. It can be concluded that closing of educational institutions and lockdown (recommendation for staying at home and moving only if necessary) are the most important exogenous factors for describing mobility, while the curfew and the mandatory use of mask in all public areas are not significant factors. In addition, seasonal models appear to produce better forecasts than the non-seasonal ones.
ID | ad110 |
Presentation | |
Full Text | |
Tags | culture, driver behaviour, machine learning, statistical modelling, urban mobility |