Concentration inequalities for sums of random variables, each having power bounded tails
Preprint
Abstract
In this work we present concentration inequalities for the sum $S_n$ of independent integer-valued not necessary indentically distributed random variables, where each variable has tail function that can be bounded by some power function with exponent $-\alpha$. We show that when $0<\alpha\leq 1$, then the sum does not have finite expectation, however, with high probability we have that $|S_n|=O\left(n^{1/\alpha}\right)$. When $\alpha>1$, then the sum $S_n$ is concentrated around its mean. Since the r.vs. that constitute the sum has tails, which can be bounded by some power function, it follows that results of this paper are applicable to a wide range of different distributions, including the exponentially decaying ones.
Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- Some Typical Properties of the Spatial Preferred Attachment Model2012
- A Critical Point for Random Graphs with a Given Degree Sequence2011
- Power-Law Distributions in Empirical DataSIAM Review, 2009
- Random 2-SAT with Prescribed Literal DegreesAlgorithmica, 2007
- Power laws, Pareto distributions and Zipf's lawContemporary Physics, 2005
- Emergence of Scaling in Random NetworksScience, 1999
- Sharp thresholds of graph properties, and the $k$-sat problemJournal of the American Mathematical Society, 1999
- The Size of the Giant Component of a Random Graph with a Given Degree SequenceCombinatorics, Probability and Computing, 1998