This work executes the price forecasting of the Mexican Crude Oil Export Blend (MME) using two methodologies. The first consists of the identification, estimation, and validation of an ARMA(p,q) time series model. An ARMA(2,0,2) model is chosen for yielding the minimum value for three information criteria (AIC, BIC, HQIC) and the Root Mean Square Error (RMSE). Secondly, the trend and volatility parameters of the Stochastic Differential Equation (SDE) of a Geometric Brownian Motion (GBM) are estimated, and predictions are executed through Monte Carlo simulation with 20,000 trajectories. According to the empirical evidence analyzed when comparing the RMSE performance measures of both models, a value of 1.27 was obtained with the ARMA(2,0,2), while a value of 35 was obtained with the GBM. In this sense, the advantages and limitations of the proposed models differ, leading to the conclusion that, for short time periods, discrete-time models are more efficient than continuous-time models.