Hausdorff moment problem: reconstruction of distributions.
Stat Probab Lett 2008 Sep; 78(12):1612-1618
The problem of approximation of the moment-determinate cumulative distribution function (cdf) from its moments is studied. This method of recovering an unknown distribution is natural in certain incomplete models like multiplicative-censoring or biased sampling when the moments of unobserved distributions are related in a simple way to the moments of an observed distribution. In this article some properties of the proposed construction are derived. The uniform and L1-rates of convergence of the approximated cdf to the target distribution are obtained.
Information-processing; Mathematical-models; Statistical-analysis; Laboratory-techniques; Models
Robert M. Mnatsakanov, Department of Statistics, West Virginia University, P.O. Box 6330, Morgantown, WV 26506
Statistics & Probability Letters