Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I am interested in a problem and I do not know where to start looking for possible similar setting. If anyone has a direction to suggest, it would be greatly appreciated.

Consider a (finite) set $\mathcal{X}$ and a set of functions $\mathcal{F}=\{f_i : \mathcal{X} \to \mathbb{R}\}_{i=1}^k$. I am interested in conditions on $\mathcal{X}$ and $\mathcal{F}$ such that there exists an $\alpha_{\mathcal{X},\mathcal{F}}\in\mathbb{R}$ for which any random variable $X\in \mathcal{X}$ has a corresponding (non-random) element $\hat{X}\in\mathcal{X}$ satisfying $$ f_i(\hat{X}) \leq \alpha_{\mathcal{X},\mathcal{F}} \mathbb{E}\{ f_i(X)\}$$ for all $i\in[k]$.

I am certainly not hoping for a solution, but just opinions on what related subjects I could look at.

share|improve this question

1 Answer 1

Consider $\mathcal F = \{ f_i : \mathbb R\rightarrow \mathbb R,\; f_i \text{ convex}, \; f_i \in L^1 \}_{i=1}^k $.

Now for all $X \in \mathbb R$ random variables such that $X^- \in L^1$ define, $\hat X = E[X]$, then by Jensen's inequality you have

$$ f_i(\hat X) \leq \mathbb E[f_i(X)] \quad \alpha_{\mathbb R, \mathcal F} = 1 $$ I guess you can play around this inequality and get different version.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.