SearcharxivSearch

arXiv subjects

Fred Torcaso

Publications and source records attributed to Fred Torcaso.

3 recordsLinked to original sources

A Geometric(1/2) Distribution Associated with Record Breaking

Let $X_i,i=0,1,\ldots$ be a sequence of iid random variables whose distribution is continuous. Associated with this sequence is the sequence $(i,X_i),i=0,1,\ldots$. Let ${\cal R}_{n}$ denote the set of Pareto optimal elements of $\{ (i,X_i):i=0,\ldots,n\}.$ We refer to the elements of ${\cal R}_{n}$ as the current records at time $n,$ and we define $R_n=\vert {\cal R}_n\vert,$ the number of such records. Observe that $R_n$ has $\{1,\ldots,n+1\}$ as its support. When $(n,X_n)$ is realized, it is a Pareto optimal element of $\{ (i,X_i)~:~i=0,\ldots,n\}$ and ${\cal R}_{n} \backslash (n,X_n) \subset {\cal R}_{n-1}.$ Then we refer to those elements of ${\cal B}_n = {\cal R}_{n-1} \backslash {\cal R}_{n}$ as the records broken at time $n.$ Let $B_n= \vert {\cal B}_n \vert.$ We show that $$P[B_n = k] \rightarrow 1/2^{k+1} \mbox{ for } k=0,1,2,\ldots.$$

math.PR

To replace or not to replace in finite population sampling

We revisit the classical result in finite population sampling which states that in equally-likely "simple" random sampling the sample mean is more reliable when we do not replace after each draw. In this paper we investigate if and when the same is true for samples where it may no longer be true that each member of the population has an equal chance of being selected. For a certain class of sampling schemes, we are able to obtain convenient expressions for the variance of the sample mean and surprisingly, we find that for some selection distributions a more reliable estimate of the population mean will happen by replacing after each draw. We show for selection distributions lying in a certain polytope the classical result prevails.

math.ST