We present a novel method for estimating physical properties of an underground hydrocarbon reservoir, on the basis of generally measured wellbore ow rate and pressure signals at the bottom of a producing well. The method uses instrumental variable-based system identification techniques to solve for a closed-loop errors-in-variables problem. It is different from the conventional methods as it allows the instrumental variable signal to be correlated with the input and output signals' noise. This property increases the number of possible candidates to be used as the instrumental variable signal. The application of the proposed method has been investigated on a synthetic case study.