COMPARISON OF THE INFORMATION-THEORETIC METHODS TO ESTIMATE THE INFORMATION FLOW IN A DYNAMICAL SYSTEM Information-theoretic quantities are widely used to quantify the relationship between different variables. Among these, mutual information is utilized to express the information that is shared between two variables. However, this quantity does not provide directionality regarding the information flow between the variables. Recently, transfer entropy has been proposed in the literature to estimate the direction of information flow in addition to its magnitude. Here, our goal is to estimate the transfer entropy from observed data accurately. Although there are various methods in the literature, there is no single recipe regarding the optimal way of estimating transfer entropy. In this work, we compare several methods along with a method that we propose, on a Lorenz model. Here, our goal is to develop the required appropriate and reliable mathematical tool synthesizing from all of these disjoint methods used in fields ranging from biomedicine to telecommunications and apply the resulting technique on tropical data sets to better understand events such as the Madden Julian Oscillation in the future.