Synthesis of statistical knowledge from time-dependent data

A general approach to analyzing multivariate time-dependent system processes with discrete-valued (both nominal and ordinal) and/or continuous-valued outcomes is presented. The approach is based on an event-covering method which selects (or covers) a subspace from the outcome space of an n-tuple of variables for estimation purposes. From the covered subspace, statistically interdependent events are selected as statistical knowledge for forecasting unknown events. The event-covering method presented is based on the use of restricted variables with only a subset of the outcomes considered. An extension to the event-covering method based on the selection of joint outcomes is discussed. The testing of this method using climatic data and simulated data which model situations in real life is described. The experiments show that the method is able to detect statistically relevant information, describe it in a meaningful and comprehensible way, and use this information for a reliable estimation (or forecast) of the missing values that will occur at some future time