We present a new, non-parametric forecasting method for data where continuous values are observed discretely in space and time. Our method, "light-cone reconstruction of states" (LICORS), uses physical principles to identify predictive states which are local properties of the system, both in space and time. LICORS discovers the number of predictive states and their predictive distributions automatically, and consistently, under mild assumptions on the data source. This leads to a natural measure of local predictive complexity, which can be used for automatic pattern discovery. Our work provides applied researchers with a new, highly automatic method to analyze and forecast spatio-temporal data. This is joint work with Georg Goerg (http://arxiv.org/abs/1206.2398)
Cosma Shalizi is an associate professor in the Department of Statistics at Carnegie Mellon University, where he is also affiliated with the Machine Learning Department and the Center for the Neural Basis of Cognition, and is an external faculty member of the Santa Fe Institute. He has a bachelor's degree from UC Berkeley and a Ph.D. from UW-Madison, both in physics. He works on non-parametrics prediction for time series and spatial dynamics, power law distributions, network modeling, Bayesian asymptotics, macroeconomic forecasting, neuroscience, causal inference and ensemble methods. He blogs athttp://bactra.org/weblog/ .