General market prediction using position specification language

摘要

Methods evaluating propositions about timeseries are provided. A service accepts a proposition about one or more timeseries, and the service monitors those timeseries. A proposition can reference multiple timeseries with different time intervals and different units. When new data is available, the service evaluates the proposition to determine if the proposition is true, false, or neither. The service automatically performs unit conversions and selects relevant, previously observed values before evaluating a proposition when new data is available.

主权项

1. A method for automatic timeseries prediction evaluation using an information system, the method comprising:
receiving a position specified by a position specification in a position specification language; extracting variable specifications from the position specification in a variable specification language; resolving each variable specification by resolving source, series, and index data to obtain a function to obtain a value with zero or more associated units; observing each variable using the function; simulating future variable observations and estimating the probability of the position obtaining based on the simulated future variable observations, wherein samples are obtained from models based on historical variable observations; coordinating an evaluation of the position by generating a sequence of tuples of values for each value such that the horizons of each tuple value intersect with all other values in that tuple; evaluating the position specification for each tuple until the position is satisfied or rejected, wherein the evaluating relates each variable specification to its corresponding tuple value, which is converted to required units by the position specification; transforming a representation of the position to reflect the result of the evaluation; and notifying the position's submitter of a change in position state.