Numerous stock market analysis methods have been proposed from simple moving average to the use of artificial intelligence such as neural networks and Bayesian networks. In this paper, we introduce a new concept and a methodology that enable predictability of asset price movement in the market by way of inference from the past data. We use schema to describe an economic instance, and a set of schema in time series to describe the flow of economic instances in the past. Within the schema, we introduce a concept of velocity and momentum to effectively characterize the dynamic nature of the market. We compare the current and the past instances to identify resemblance and take inference as a predictive capability of future asset price movement.