Probability for Statistics and Machine Learning pp 463-504 | Cite as

# Discrete Time Martingales and Concentration Inequalities

## Abstract

For an independent sequence of random variables *X* _{1}, *X* _{2}, *…*, the conditional expectation of the present term of the sequence given the past terms is the same as its unconditional expectation. Martingales let the conditional expectation depend on the past terms, but in a special way. Thus, similar to Markov chains, martingales act as natural models for incorporating dependence into a sequence of observed data. But the value of the theory of martingales is much more than simply its modeling value. Martingales arise, as natural byproducts of the mathematical analysis in an amazing variety of problems in probability and statistics. Therefore, results from martingale theory can be immediately applied to all these situations in order to make deep and useful conclusions about numerous problems in probability and statistics. A particular modern set of applications of martingale methods is in the area of concentration inequalities, which place explicit bounds on probabilities of large deviations of functions of a set of variables from their mean values. This chapter gives a glimpse into some important concentration inequalities, and explains how martingale theory enters there. Martingales form a nearly indispensable tool for probabilists and statisticians alike.

## Keywords

Maximal Inequality Sequential Probability Ratio Test Concentration Inequality Martingale Theory Fair Game## References

- Azuma, K. (1967). Weighted sums of certain dependent random variables,
*Tohoku Math. J*., 19, 357–367.MathSciNetzbMATHCrossRefGoogle Scholar - Brown, B.M. (1971). Martingale central limit theorems,
*Ann. Math. Statist*., 42, 59–66.MathSciNetzbMATHCrossRefGoogle Scholar - Burkholder, D.L. (1973). Distribution function inequalities for martingales,
*Ann. Prob*., 1, 19–42.MathSciNetzbMATHCrossRefGoogle Scholar - Burkholder, D.L., Davis, B., and Gundy, R. F. (1972). Integral inequalities for convex functions of operators on martingales,
*Proc. Sixth Berkeley Symp. Math. Statist. Prob*., Vol. II, 223–240, University of California Press, Berkeley.Google Scholar - Chow, Y.S. and Teicher, H. (2003).
*Probability Theory: Independence, Interchangeability, Martingales*, Springer, New York.zbMATHGoogle Scholar - Chung, K.L. (1974).
*A Course in Probability*, Academic Press, New York.zbMATHGoogle Scholar - DasGupta, A. (2010).
*Fundamentals of Probability: A First Course*, Springer, New York.zbMATHCrossRefGoogle Scholar - Davis, B. (1970). On the integrability of the martingale square function,
*Israel J. Math*., 8, 187–190.MathSciNetzbMATHCrossRefGoogle Scholar - Devroye, L. (1991).
*Exponential Inequalities in Nonparametric Estimation, Nonparametric Functional Estimation and Related Topics*, 31–44, Kluwer Acad. Publ., Dordrecht.Google Scholar - Doob, J.L. (1971). What is a martingale?,
*Amer. Math. Monthly*, 78, 451–463.MathSciNetzbMATHCrossRefGoogle Scholar - Fristedt, B. and Gray, L. (1997).
*A Modern Approach to Probability Theory*, Birkh*ä*user, Boston.Google Scholar - Hall, P. and Heyde, C. (1980).
*Martingale Limit Theory and Its Applications*, Academic Press, New York.Google Scholar - Heyde, C. (1972). Martingales: A case for a place in a statistician’s repertoire,
*Austr. J. Statist*., 14, 1–9.MathSciNetzbMATHCrossRefGoogle Scholar - Hoeffding, W. (1963). Probability inequalities for sums of bounded random variables,
*J. Amer. Statisto Assoc*., 58, 13–30.MathSciNetzbMATHCrossRefGoogle Scholar - Karatzas, I. and Shreve, S. (1991).
*Brownian Motion and Stochastic Calculus*, Springer, New York.zbMATHGoogle Scholar - Karlin, S. and Taylor, H.M. (1975).
*A First Course in Stochastic Processes*, Academic Press, New York.zbMATHGoogle Scholar - McDiarmid, C. (1989).
*On the Method of Bounded Differences, Surveys in Combinatorics*, London Math. Soc. Lecture Notes, 141, 148–188, Cambridge University Press, Cambridge, UK.Google Scholar - Williams, D. (1991).
*Probability with Martingales*, Cambridge University Press, Cambridge, UK.zbMATHCrossRefGoogle Scholar