Résumé
Russia, one of the major oil-exporting countries, is very vulnerable to fluctuations in the price of oil. This thesis aims to explore the impact of the oil price on the performance of the Russian economy described by several variables (industrial production, unemployment rate, inflation, RTS stock index, and ruble).To achieve this goal, we will use econometric methodologies in both a linear and a non-linear framework. We will start by testing the presence of linear cointegration relation and/or estimating VAR models. If necessary, we will resort to threshold cointegration to model the nonlinearities of the relation between the economic variables mentioned above.Thus, in the case of detection of a threshold cointegration, a threshold VECM model makes it possible to take into account the thresholds and the presence of several regimes. The coefficients of the error terms of the threshold VECM models inherent in each regime can better explain the fit to the equilibrium relationship compared to conventional single-regime VECM models. On the other hand, regime change models allow detection of the different regimes present in the structure of the analyzed variables; the transition from one regime to another is described by a probability matrix. Finally, wavelet analysis, which analyzes the behavior of variables in a time-frequency domain, allows us a more detailed analysis of the complexity and non-linearity of the links that may exist between the different studied variables.Thanks to all these methodologies, we were able to carry out an in-depth analysis of the Russian economy and better understand the relationships and interdependencies that may exist between the variables describing the performance of this economy.