5
$\begingroup$

Suppose that one knows how to generate (independent) random samples $X_1, X_2, \ldots$ distributed as the random varable $X$ with $\mathbb{E}[X]=\mu \in \mathbb{R}$. It is then easy to construct an unbiased estimator of the quantity $e^{\mu}$. Indeed, it suffices to generate an integer random variables $N$ such that $\mathbb{P}[N \geq k] = \frac{1}{k!}$. The random variable $$Y=1 + X_1 + X_1X_2 + \ldots + \prod_{k \leq N-1} X_k + \prod_{k \leq N} X_k$$ satisfies $\mathbb{E}[Y]=e^{\mu}$.

Question: is it possible to construct an unbiased estimator of $e^{\mu}$ that is positive with probability $1$. If the random variable $X$ is lower bounded by a constant $C$ one can Taylor expand $\exp(C) \cdot \exp(x-C)$ and use the same idea as above. What about the case where $X$ is not lower bounded? One could try to Taylor expand $\exp(m) \cdot \exp(x-m)$ where $m = \min(X_1, \ldots, X_N)$ but in this case the estimator does not satisfy $\mathbb{E}[Y]=e^{\mu}$ anymore due to the dependence between the random variable $m$ and $(X_1, \ldots, X_N)$.

$\endgroup$
6
  • $\begingroup$ I would like to seek some clarification on this problem. First of all, the use of unbiased estimator is a little unconventional here. Usually, one thinks of (and defines!) an unbiased estimator as a sequence of functions for each fixed $n$ such that $\mathbb E Y_n = e^\mu$. Here, instead we consider only a single random $N$ for the entire sequence. It is then not much of a leap to ask if we can consider using the entire sequence to construct the estimator, but then your problem (quite trivially) has an affirmative answer. $\endgroup$
    – cardinal
    Jun 24, 2012 at 15:07
  • $\begingroup$ So, could you please specify the (additional) constraints you seek to place on the problem? (Note also, it is interesting that the distribution of $N$ has both its mode and median at 1 and a mean value of $e$, so we don't normally use very many elements in the sequence.) $\endgroup$
    – cardinal
    Jun 24, 2012 at 15:08
  • 1
    $\begingroup$ to put it another way: I know how to simulate random samples from the random variable $X$ that has $\mu$ as mean (it takes a finite amount of computer time for each draw) and I would like to generate a random sample from a random variable Y that has $e^\mu$ as a mean. The cost of generating the random variable $Y$ must take a finite amount of computer time (on average). $\endgroup$
    – Alekk
    Jun 24, 2012 at 22:37
  • $\begingroup$ It has recently been proved that such an algorithm cannot be constructed. $\endgroup$
    – Alekk
    Jul 30, 2013 at 5:17
  • $\begingroup$ @Alekk could you share with us the proof or reference to it? $\endgroup$
    – Memming
    Jul 17, 2014 at 17:26

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.