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.

One can decompose a $\sigma$-finite measure $\mu$ on $\mathbb{R}$ in three parts:

$\mu_{ac}$: absolutely continuous

$\mu_{sc}$: singular continuous

$\mu_{pp}$: pure point

A common example for a singular continuous probability measure is Cantor's function as cdf. Such a cdf is continuous. I have two question:

(1) Do singular cont. probability measures come up? E.g. as law of pure jump L\'evy processes, for which there are criteria to guarantee a density. What about semimartingales?

(2) The characteristic function can be defined in two ways $\int e^{itx}dF(x)$ or $\int e^{itx}d\mu_{sc}(x)$. Though there is no density function, can $\int e^{itx}dF(x)$ be approximated by a sequence of, say a continuous $\mathcal{L}^1$ functions, such that $\int e^{itx}dF(x) = \lim_{n\rightarrow\infty}\int e^{itx}f_n(x)dx$?

I can't find anything on that, so maybe singular measures are quite opaque objects in probability.

share|improve this question
add comment

Your Answer

 
discard

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

Browse other questions tagged or ask your own question.