Many pattern matching approaches have been applied in financial time series to detect chart patterns and predict price trends. In this thesis, we propose an extended hidden semi-Markov model for chart pattern matching. In our approach, a hidden semiMarkov model is trained in a new approach and a Viterbi algorithm is used to detect chart patterns. We compare the proposed model with current approaches on a set of templates selected from 53 chart patterns. Experiments involving datasets of synthetic chart patterns showed that the proposed approach had a higher accuracy and recall than other pattern matching approaches. Experiments were also conducted on a real dataset comprising the historical prices of several stocks. In addition, we evaluate the effectiveness and accuracy of four approaches to financial time series pattern matching when used with four segmentation methods.