System and method for predicting significant events using a progress curve model

摘要

Described is system for predicting significant events using a progress curve model. The system first determines Z-score values for a predetermined period of a time series to generate a Z-score time series. The Z-score time series are partitioned into a plurality of E-periods to define time frames for progress curve model (PCM) fitting. An E-period is defined as a period of escalation and de-escalation. Finally, a future event is predicted based on an absolute Z-score value that is greater than or equal to a predetermined number.

主权项

1. A system for predicting events using a progress curve model, the system comprising:
one or more processors and a non-transitory memory having instructions encoded thereon such that when the instructions are executed, the one or more processors perform operations of: determining Z-score values for a predetermined period of a time series; generating a Z-score time series from the Z-score values; partitioning the Z-score time series into a plurality of E-periods to define a plurality of time frames for progress curve model (PCM) fitting, wherein an E-period is defined as a period of escalation and de-escalation; deriving a set of regression parameters for PCMs; and generating a prediction of a financial market event in a current E-period based on the set of regression parameters.