Modeling the dynamic response of automobile sales in troubled times: a real-time Vector Autoregressive analysis with causality testing for Greece

In this paper, we investigate the factors that affect multi-segments automobile sales in Greece. Various relevant quantitative techniques have been employed, such as stationarity, causality and cointegration. A Vector Autoregressive (VAR) model was also developed and long-term impacts of the different variables of interest on car sales have been estimated through generalized impulse response functions (GIRF). The impact of the current financial crisis on the Greek automobile market was also taken into account. The results show that fuel prices Granger cause total car sales. The results also indicate the absence of long run cointegrating relationships among the variables. The full blown model shows that demand for new automobiles depends on the existing social, financial and political conditions of the local economy and that the various shocks observed have a temporary medium-run character on car sales, whereas the system is found to be stable.