In this paper, we introduce a class of generalized Browaian sheets and extend the OUP; in [1]. By studying some fundamental properties of the former process, we first derive some of simple relaxed past Markov properties of the rwo processes. Next, utilizing the relation between the two processes, we successively prove that, for two kinds of general sets, they possess the general relaxed past Markov properties and the Levy Markoy properties re-spectively. As a result, we conclude that the OUP; in [1] also has general Markov properties for the corresponding sets.

This paper extends the 4m2n minimum aberration designs (MA designs) of Wu and Zhang (1993) to the case of (S2)Sn−k, where S is any prime or prime power. Some basic properties of (S2)Sn−k MA designs, including the relations with Sn−k MA designs, are discussed. The (9)3n−k MA designs with 27 runs and (16)4n−k MA designs with 64 runs are tabulated, and some (4)2n−k MA designs are constructed using the above relations.

Based on the work of Wang and Fang (1990a, b), this paper extends the def-inition of discrepancy of a set of points with respect to a distribution. To construct multivariate uniform designs, two transformations, discrepancy-preserving transfor-mation and density-preserving transformation, and a method of separating variables are introduced. Also, for some special domains useful in statistics, some related transformations are given. In particular, as an application of the method of sepa-rating variables, we give an approach of constructing a uniform design in the Stiefel manifold and outline its applications in projection pursuit methods.

this paper, the only assumptions on the distribution of data are those concerning rst two moments. Our purpose is to estimate the parameter of interest in the presence of nuisance parameter under these weak assumptions on the distribution. We de ne a quasi-least favorable curve andconstruct its estimator, andthen yielda pro le quasi-score function of the parameter of interest. The estimator of parameter of interest obtainedfrom this score function is asymptotically e1cient. On the other hand, we employ this method to estimate the parameter in the semiparametric model. In this model the nonparametric component plays the role of nuisance parameter andit takes values in a in nite-dimensional space. The methodis also available for semiparametric model andthe estimator obtainedby the extension is asymptotically e1cient.

In this paper formulas for computing the E (s2) values of some kinds of E (s2) optimal super-saturated designs are given, and a general algorithm of constructing E (s2) optimal supersaturated designs from cyclic BIBDs is proposed. Within this class of designs, by further discriminating the pairwise correlations, e cient designs of runs from 6 to 24 are constructed and tabulated. Comparisons with other existing designs are made at last, demonstrating the e ectiveness of our method.