arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 0905.2944math.PRmath.STstat.TH
收藏

密度卷积的指数亏度

Exponential deficiency of convolutions of densities

Iosif Pinelis

摘要:

如果概率密度p(\x)(\x \in \R^k)是有界的,并且R(t):= \int \exp(t \ell(\x))\d \x < \infty对于某些线性泛函\ell和所有t \in(0,1),那么,对于每个t \in(0,1)和所有足够大的n,t-倾斜密度p_t(\x):= \exp(t \ell(\x))p(\x)/R(t)的n重卷积是有界的。这是一个普遍的推论,“非独立身份证”。“的结果,这也被证明享有一定的最优性属性。这样的结果是有用的鞍点近似。

英文摘要:

If a probability density p(\x) (\x\in\R^k) is bounded and R(t) := \int \exp(t\ell(\x)) \d\x < \infty for some linear functional \ell and all t\in(0,1), then, for each t\in(0,1) and all large enough n, the n-fold convolution of the t-tilted density p_t(\x) := \exp(t\ell(\x)) p(\x)/R(t) is bounded. This is a corollary of a general, "non-i.i.d." result, which is also shown to enjoy a certain optimality property. Such results are useful for saddle-point approximations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑