TinTO: A Tool for View-Based Analysis of Stock Market Data Streams

TinTO is an experimental system aiming at demonstrating the usefulness and feasibility of incrementally evaluated SQL queries for analyzing a wide spectrum of data streams. As application area we have chosen the technical analysis of stock market data, mainly because this kind of application exhibits sufficiently many of those characteristics for which relational query technology can be reasonably considered in a stream context. TinTO is a technical investor tool for computing so-called technical indicators, numerical values calculated from a certain kind of stock market data, characterizing the development of stock prices over a given time period. Update propagation is employed for the incremental recomputation of indicator views defined over a stream of continuously changing price data. 1 Technical Analysis of Stock Market Data Technical analysis (TA) is concerned with the prediction of future developments of stock market prices. TA uses so-called technical indicators, numerical values derived from the past development of prices of a certain stock. In principle, indicators are functions applied to the price history of a certain stock and a point in time. A technical analyst is usually interested in the change of indicator values over a certain time period in order to automatically derive buy or sell signals for stocks. The stream of time-stamped price data is an ordered append-only stream provided by a stock market such as the New York stock exchange. In general, indicator computation is based on a division of a stocks price history into consecutive time intervals ti of equal length. Interval length is user-defined and usually ranges from 1 second to as long as one year. For each time interval, the mean average is determined representing the typical price TP of stock s with respect to ti. A technical indicator definition is based on these values and an additional user-defined parameter n specifying the number of consecutive time intervals under consideration. As an example, consider the commodity channel index (CCI) which is an overbought/oversold indicator, whose value typically oscillates around zero: CCIn(s, ti) := TP(s,ti)− ∑n−1 l=0 TP(s,ti−l) n