1. Suppose we have a sequence of independent random vars $X_n$ and consider a sequence of truncated random variables $Y_n=X_n1_{X_n\le n}$ s.t. $E[Y_n]=0$.

  2. We know that $X_n$'s and $Y_n$'s are convergence equivalent:

$$\sum_{n=1}^\infty P(X_n\ne Y_n)<\infty$$

  1. $\sum X_n$ converges to a nondegenerate random variable $X$.

  2. Finally, for any $\epsilon>0$, $P(|\sum_{m=k}^\infty(X_m-Y_m)|>\epsilon)\rightarrow 0$ as $k\rightarrow \infty$.

In this settings we can say that for $\epsilon>0$ there exists $k$ s.t.

$$P(\sum_{m=k}^\infty X_m\le x)-\epsilon\le P(\sum_{m=k}^\infty Y_m\le x)\le P(\sum_{m=k}^\infty X_m\le x)+\epsilon$$

However, can we give a more precise bound on $P(\sum_{m=k}^\infty Y_m\le x)$. Especially, I'm interested in comparing the moments of limiting distributions.



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.