ISSN:0233-1888    

DOI:10.1080/02331889108802307

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

The problem of combining linear and ellipsoidal restrictions in linear regression is investigated. Necessary and sufficient conditions for compactness of the restriction set are proved assuring the existence of a minimax estimator. When the restriction set is not compact a minimax estimator may still exist for special loss functions arid regression designs

ISSN:0233-1888    

DOI:10.1080/03610929108802299

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

The “flat” or “saddle point” approximation of the probability density of the M.L. estimator is used to derive pivotal quantities which are “almost exact” in a nonlinear model y=η(θ)+∈, ∈∼N(0, σ2H). This allows to construct “almost exact” confidence regions for θ in the case when σ is unknown

ISSN:0233-1888    

DOI:10.1080/02331889108802300

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

ISSN:0233-1888    

DOI:10.1080/02331889108802301

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

Existence, uniqueness, and numerical computations of the ML-estimator in linear regression models with stationary AR(p) errors are discussed. The ML-estimator can be computed as solution of a nonlinear optimization problem without constraints but with a special stepsize control

ISSN:0233-1888    

DOI:10.1080/02331889108802302

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

In the paper minimax estimators of the difference of multinomial parameters and multivariate hypergeometric are determined for quadratic loss function. Solved is also problem of minimax estimation of the difference of means of bounded random varia-bles and also the parameterof multivariate hypergeometric distribution when ivfi,…, Mrs N are unknown

ISSN:0233-1888    

DOI:10.1080/02331889108802303

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

The VON MISES method is generalized to the nonparametric case in order to prove a triangular version of the √n-consistency for an estimatorwhereis a probability distribution on a general metric space, P≗nits empirical estimator and φ a HADAMARD-differentiable function, Convergence results of bootstrap procedures are deduced for bivariate survival data.

ISSN:0233-1888    

DOI:10.1080/02331889108802304

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

We develop Edgeworth expansions for kernel density estimators, and for their bootstrap counterparts. Those results are applied to the bootstrap confidence interval problem for densities, paying particular attention to the issue of bias. Two approaches to bias correction are considered: explicit bias estimation, and deliberate undersmoothing to render bias negligible, The latter method is shown to have advantages over the former from the viewpoint of coverage accuracy of onesided confidence intervals. Addition-ally, it produces shorter two-sided intervals

ISSN:0233-1888    

DOI:10.1080/02331889108802305

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

The paper deals with testing of periodicity for time series with missing obser-vations. Two schemes of missing observations are considered: regularly missing obser-vation and observations missing randomly according to the BERNOULLI model

ISSN:0233-1888    

DOI:10.1080/02331889108802306

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

This paper suggests several tests for checking the nullhypothesis that a point pattern observed in a (large) rectangular sampling window D⊂Rdbelongs to a realization of a stationary Poisson Process based on test statistics measuring the distance between the empirical and the true second moment functionvolume of the unit sphere im Rd. We consider the case of known intensity and the case of estimated intensity separately. On the basis of weak convergence of the corresponding diviation processes on [0,R] (as D becomes large) to Gaussian Processes various analogues to the classical goodness-of-fit tests X2Kolmogorov-SMIRNOV, CRAMER-VON MISES test) are developed and discussed. Tabulated values of the test statistics based on simulated data compared with the quantiles of their limit distributions illustrate the applicability of the proposed tests

ISSN:0233-1888    

DOI:10.1080/02331889108802308

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor