2011/05/25 by Haim H. Permuter, Young-Han Kim, Tsachy Weissman · 3 citations
Computer Science · Decision Sciences · #Computability, Logic, AI Algorithms #Advanced Bandit Algorithms Research #Machine Learning and Algorithms
paper · doi:10.1109/tit.2011.2136270
openalex publication_date 2011/05/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/29
We investigate the role of directed information in portfolio theory, data compression, and statistics with causality constraints. In particular, we show that directed information is an upper bound on the increment in growth rates of optimal portfolios in a stock market due to causal side information. This upper bound is tight for gambling in a horse race, which is an extreme case of stock markets. Directed information also characterizes the value of causal side information in instantaneous compression and quantifies the benefit of causal inference in joint compression of two stochastic processes. In hypothesis testing, directed information evaluates the best error exponent for testing whether a random processYcausally influences another processXor not. These results lead to a natural interpretation of directed informationI(Yn→Xn) as the amount of information that a random sequenceYn= (Y1,Y2,...,Yn) causally provides about another random sequenceXn= (X1,X2,...,Xn). A new measure, directed lautum information, is also introduced and interpreted in portfolio theory, data compression, and hypothesis testing.