Bayesian approach to prediction and the spacings in the exponential distribution

Summary A series of independent samples are drawn from a general population with positive variationf(x,ϕ), x>0. Based on the Bayesian approach, a general predictive distribution is given, to predict a statistic in the future sample based on the statistics in the earlier samples (or stages). Few general classes of distributions of this type like Koopman-Pitman family, power function family and Burr's class of distributions are considered to show how this procedure works in predicting order statistics in the future sample. Also, the sum of the spacings in the future samples from an exponential population is predicted in terms of similar sum of spacings in the earlier samples. Discussion on the variance of this predictive distribution is dealt with. Finally, an illustrative example with simulated samples from an exponential population gives actual prediction of an order statistic as well as the sum of spacings in the future sample.     

Regressions and monotone chains: A ramsey-type extremal problem for partial orders

Aregression is a functiong from a partially ordered set to itself such thatg(x)≦x for allz. Amonotone k-chain is a chain ofk elementsx ₁<x ₂ <...<x _k such thatg(x ₁)≦g(x ₂)≦...≦g(x _k ). If a partial order has sufficiently many elements compared to the size of its largest antichain, every regression on it will have a monotone (k + 1)-chain. Fixingw, letf(w, k) be the smallest number such that every regression on every partial order with size leastf(w, k) but no antichain larger thanw has a monotone (k + 1)-chain. We show thatf(w, k)=(w+1) ^k . Dedicated to Paul Erdős on his seventieth birthday Research supported in part by the National Science Foundation under ISP-80-11451.     

Estimation with Sequential Order Statistics from Exponential Distributions

The lifetime of an ordinary k-out-of-n system is described by the (n–k+1)-st order statistic from an iid sample. This set-up is based on the assumption that the failure of any component does not affect the remaining ones. Since this is possibly not fulfilled in technical systems, sequential order statistics have been proposed to model a change of the residual lifetime distribution after the breakdown of some component. We investigate such sequential k-out-of-n systems where the corresponding sequential order statistics, which describe the lifetimes of these systems, are based on one- and two-parameter exponential distributions. Given differently structured systems, we focus on three estimation concepts for the distribution parameters. MLEs, UMVUEs and BLUEs of the location and scale parameters are presented. Several properties of these estimators, such as distributions and consistency, are established. Moreover, we illustrate how two sequential k-out-of-n systems based on exponential distributions can be compared by means of the probability P(X < Y). Since other models of ordered random variables, such as ordinary order statistics, record values and progressive type II censored order statistics can be viewed as sequential order statistics, all the results can be applied to these situations as well.     

Intersections ofk-element sets

LetF be a collection ofk-element sets with the property that the intersection of no two should be included in a third. We show that such a collection of maximum size satisfies .2715k+o(k)≦≦log₂ |F|≦.7549k+o(k) settling a question raised by Erdős. The lower bound is probabilistic, the upper bound is deduced via an entropy argument. Some open questions are posed. This research has been supported in part by the Office of Naval Research under Contract N00014-76-C-0366. Supported in part by a NSF postdoctoral Fellowship.     

On the variation of co-ordinates in subspaces

Summary Let1≦k≦n−1, 2≦n. This paper examines the vectors δ_p(L_k), where L_k is a k-dimensional subspace of an n-dimensional space, and the co-ordinates of δ_p(L_k) are given below by (1.1). For fixed k, the set of such vectors as L_k varies is determined for p=2. For general p, information is given on upper and lower bounds for the sum of the co-ordinates of δ_p(L_k). Dedicated to the sixtieth birthday of Prof. Edgar R. Lorch This research was supported in part by the Office of Naval Research under contract number Nonr 3775(09), NR 047040.     

Extreme value limit laws in the nonidentically distributed case

LetX ₁, ...,X _n be independent random variables, letF _i be the distribution function ofX _i (1≦i≦n) and letX _1n ≦... ≦X _nn be the corresponding order statistics. We consider the statisticsX _kn, wherek=k(n),k/n → 1 andn−k → ∞. Under some additional restrictions concerning the behaviour of the sequences {a _n>0,b _n,k(n),F _n} we characterize the class of all distribution functionsH such that Prob{(X _kn −b _n )/a _n <x)}→H. Dedicated to the Memory of N. V. Smirnov (1900–1966)     

On a problem of klee concerning convex polytopes

A non-simpliciald-polytope is shown to have strictly fewerk-faces ([(d−1)/2]≦k≦d−1) then some simpliciald-polytope with the same number of vertices; actual numerical bounds are given. This provides a strong affirmative answer to a problem of Klee.     

The largest projections of regular polytopes

The projections of the regularn-dimensional simplex and crosspolytope intoR ^k with the largestk-volume are determined here for the casesk=2,n≧2 andk=3, 4≦n≦6. The proofs involve a combination of exterior algebra and computer gradient methods.     

Interior points of convex hulls

If a setX ⊂E ⁿ has non-emptyk-dimensional interior, or if some point isk-dimensional surrounded, then the classic theorem of E. Steinitz may be extended. For example ifX ⊂E ⁿ has int _k X ≠ 0, (0 ≦k≦n) and ifp ɛ int conX, thenp ɛ int conY for someY ⊂X with cardY≦2n−k+1.     

Selection of the number of regression variables; A minimax choice of generalized FPE

Summary A generalized Final Prediction Error (FPE_α)_ criterion is considered. Based onn observations, the numberk of regression variables is selected from a given range 0≦k≦K, so as to minimize . It is shown that if α tends to infinity withn, the selection is consistent but the maximum of the mean squared error of estimates of parameters diverges to infinity with the same order of divergence as that of α. A meaningful minimax choice of α exists for a regret type mean squared error, while for simple mean squared error it is trivially 0. The minimax regret choice of α converges to a constant, approximately 3.5 forK≧8 ifn−K increases simultaneously withn, otherwise it diverges to infinity withn.     

Order restricted inference for sequential k-out-of-n systems

Sequential order statistics have been introduced to model sequential k-out-of-n systems which, as an extension of k-out-of-n systems, allow the failure of some components of the system to influence the remaining ones. Based on an independent sample of vectors of sequential order statistics, the maximum likelihood estimators of the model parameters of a sequential k-out-of-n system are derived under order restrictions. Special attention is paid to the simultaneous maximum likelihood estimation of the model parameters and the distribution parameters for a flexible location-scale family. Furthermore, order restricted hypothesis tests are considered for making the decision whether the usual k-out-of-n model or the general sequential k-out-of-n model is appropriate for a given data.     

Selection from poisson processes

Summary There are givenk Poisson processes with mean arrival times 1/λ₁,...1/λ _k . Let λ_[1]≦λ_[2]≦...≦λ_[k] denote the ordered set of values λ₁...,λ_[k]. We consider three procedures for selecting the process corresponding to λ_[k]. The processes are observed until there areN arrivals from any of the given processes, when the processes are observed continuously, or until there are at leastN arrivals, when the processes are observed at successive intervals of time whereN is a pre-determined positive integer. In the continuous case, the process for which theNth arrival time is shortest, is selected. In the discrete case, the selection involves certain randomization. Given (λ_[k]/λ_[k-1])≧0>1, it is shown that the probability of a correct selection (Pcs) is minimized whenθλ_[1]=θλ_[2]=...=θλ_[k-1]=θλ_[k]=θλ, say. The Pcs for this configuration is independent of λ for two of the given procedures, and monotone increasing in λ for the third. The value ofN is determined by a lower bound placed on the value of the Pcs. The problem of selecting from given Poisson processes for the discrete case is related to the problem of selecting from given Poisson populations. An application of the given procedures to a problem of selecting the “most probable event” from a multinomial population, is considered.     

Prediction of multivariate time series by autoregressive model fitting

Suppose the stationary r-dimensional multivariate time series {y_t} is generated by an infinite autoregression. For lead times h≥1, the linear prediction of y_t+h based on y_t, y_t−1,… is considered using an autoregressive model of finite order k fitted to a realization of length T. Assuming that k → ∞ (at some rate) as T → ∞, the consistency and asymptotic normality of the estimated autoregressive coefficients are derived, and an asymptotic approximation to the mean square prediction error based on this autoregressive model fitting approach is obtained. The asymptotic effect of estimating autoregressive parameters is found to inflate the minimum mean square prediction error by a factor of (1 + kr/T).     

A property of (v,k, λ)-designs

It is shown that if the subsetsX ₁,...,X _v of a setX form a (v, k, λ)-design, then there does not exist another subsetX _v+1 ofX havingany cardinalityk ₁ and intersecting each of theX _j, 1≦j≦v, inany number λ₁ of elements, where 0<k ₁<v and 0<λ₁<k (in order to avoid uninteresting cases).     

Packing nearly-disjoint sets

De Bruijn and Erdős proved that ifA ₁, ...,A _k are distinct subsets of a set of cardinalityn, and |A _i ∩A _j |≦1 for 1≦i<j ≦k, andk>n, then some two ofA ₁, ...,A _k have empty intersection. We prove a strengthening, that at leastk /n ofA ₁, ...,A _k are pairwise disjoint. This is motivated by a well-known conjecture of Erdőds, Faber and Lovász of which it is a corollary. Partially supported by N. S. F. grant No. MCS—8103440     

On sperner families in which nok sets have an empty intersection III

LetR be anr-element set and ℱ be a Sperner family of its subsets, that is,X ⊈Y for all differentX, Y ∈ ℱ. The maximum cardinality of ℱ is determined under the conditions 1)c≦|X|≦d for allX ∈ ℱ, (c andd are fixed integers) and 2) nok sets (k≧4, fixed integer) in ℱ have an empty intersection. The result is mainly based on a theorem which is proved by induction, simultaneously with a theorem of Frankl.     

The analogue of eberhard’s theorem for 4-valent graphs on the torus

It is shown that every sequence {p _k | 3≦k ≠ 4} (with one exception) that satisfies the right equation is realizable by a 4-valent graph on the torus. Presented to the 15th Ontario Mathematical Meeting, March 14, 1970.     

Subset selection procedures for restricted families of probability distributions

In this paper we are interested in studying multiple decision procedures fork (k≧2) populations which are themselves unknown but which one assumed to belong to a restricted family. We propose to study a selection procedure for distributions associated with these populations which are convex-ordered with respect to a specified distributionG assuming that there exists a best one. The procedure described here is based on a statistic which is a linear function of the firstr order statistics and which reduces to the total life statistics whenG is exponential. The infimum of the probability of a correct selection and an asymptotic expression for this probability are obtained using the subset selection approach. Some other properties of this procedure are discussed. Asymptotic relative efficiencies of this rule with respect to some selection procedures proposed by Barlow and Gupta [3] for the star-ordered distributions and by Gupta [8] for the gamma populations with known shape parameters are obtained. A selection procedure for selecting the best population using the indifference zone approach is also studied. This research was supported by the Office of Naval Research Contract N00014-75-C-0455 at Purdue University. Reproduction in whole or in part is permitted for any purpose of the United States Government. Ming-Wei Lu is now at the Department of Vital and Health Statistics, Michigan.     

Dimensions of fixed point sets of involutions

Suppose the fixed point set F of a smooth involution T:M → M on a smooth, closed and connected manifold M decomposes into two components Fⁿ and F² of dimensions n and 2, respectively, with n > 2 odd. We show that the codimension k of Fⁿ is small if the normal bundle of F² does not bound; specifically, we show that k≦ 3 in this case. In the more general situation where F is not a boundary, n (not necessarily odd) is the dimension of a component of F of maximal dimension and k is the codimension of this component, and fixed components of all dimensions j, 0≦ j≦ n, may occur, a theorem of Boardman gives that . In addition, we show that this bound can be improved to k≦ 1 (hence k = 1) for some specific values of n and some fixed stable cobordism classes of the normal bundle of F² in M; further, we determine in these cases the equivariant cobordism class of (M, T). Received: 25 August 2005     

Universal PI-algebras and algebras of generic matrices

Let Ω[ξ] denote the polynomial algebra (with 1) in commutative indeterminates {ie65-1}, 1 ≦i, j ≦n, 1 ≦k < ∞, over a commutative ring Ω. Thealgebra of generic matrices Ω [Y] is defined to be the Ω-subalgebra ofM _n (Ω[ξ]) generated by the matricesY _k=({ie65-2}), 1 ≦i, j ≦n, 1 ≦k < ∞. This algebra has been studied extensively by Amitsur and by Procesi in particular Amitsur has used it to construct a finite dimensional, central division algebra Ω (Y) which is not a crossed product. In this paper we shall prove, for Ω a domain, that Ω(Y) has exponentn in the Brauer group (Amitsur may already know this fact); consequently, for Ω an infinite field andn a multiple of 4, iff(X ₁, …,X _m) is a polynomial linear in all theX _i but one (similar to Formanek’s central polynomials for matrix rings) andf ² is central forM _n (Ω), thenf is central forM _n (Ω). (The existence of a polynomial not central forM _n (Ω), but whose square is central forM _n(Ω) is equivalent to every central division algebra of degreen containing a quadratic extension of its center; well-known theory immediately shows this is the case of 4‖n and 8χn.) Also, information is obtained about Ω(Y) for arbitary Ω, most notably that the Jacobson radical is the set of nilpotent elements. Partial support for this work was provided by National Science Foundation grant NSF-GP 33591.