This research proposes a numerical approach in estimating the trend of behavior of this market. This approach is applied to a model that is inspired by catalytic chemical model, in terms of differential equations, on four composite indices, New York Stock Exchange, Hong Kong Hang Seng, Straits Times Index, and Jakarta Stock Exchange, as suggested by Caetano and Yoneyama (2011). The approach is used to minimize the difference of estimated indices based on the model with respect to the actual data set. The result shows that the estimation is able to capture the trend of behavior in stock market well.