Let $(X_1,X_2,\cdots)$ be a sequence of independent and identically distributed random variables taking values in some general normed space $(V,||\cdot||)$. Denote $\mu=E[X_1]$ and $S_n=\frac{1}{n}[X_1+\cdots+X_n]$.

This question is about the tail probability $Pr[||S_n-\mu||\geq\epsilon]$ for some $\epsilon > 0$ and I am especially concerned with the question if the tail probability decays to exponentially fast in $n$.

Is $||X_i||\leq\tau$, $\tau \in (0,\infty)$ sufficient for Hoeffdings's Inequallity to hold with $$ Pr[||S_n-\mu||\geq\epsilon]\leq \exp\left(-n^2\epsilon^2c \right), $$ where $c>0$ is some finite constant?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.