A critique towards real business cycles (RBC) modelling is their arbirtray primary asumtions and non-testability. In response, the robustness of simulation should be tested with respect to changing the assumptions, or in other words, their compatibility should be tested with regards to real data moments. In this study, we calibrate a general equilibrium model with two sources of uncertainties, productivity and oil income, for the time period of 1988-2012 (before international sanctions) and simulate it using Dynare (MatLab). Propagation mechanisms include the auto regressive structure of uncertainties plus the investment. As first practice, simulated moments of the model is compared with the real ones in Iran. In another practice, the oil is excluded, and results are compared with the non-oil sector of Iran. Besides, we check for the best filter among two selective varieties of high-frequency filters and middle-pass filters. Among filsteres, the high-frequency ones are a better separatore of cycles vs. trend for Iran; noting that the high-frequency ones obtain more noicy cycles. In modeling the non-oil sector of Iran's economy, even though the non-oil time series of the national account are employed, excluding oil from modeling makes it less reliable, probably because the volatilities originated from oil is propagated into all sectores of the economy including non-oil sectors. In sum, our findings show that first, a high frequency filter with more noisy cycles is more appropriate; and second, macro modeling must necessarily includes the oil secotre.