Modeling Time-Series of Stock Price by Stochastic Context-Free Grammars

We propose a stochastic context-free grammar model whose structure can al-ternatively be viewed as a graphical model, and use it to model time-series of stock prices. A method is presented for extracting characteristic patterns from a time se-ries using such kind of model. We use a modification of expectation-maximization algorithm to estimate the model parameters and present empirical findings