A multidimensional market analysis method using level-velocity-momentum time-series vector space

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.