摘要:

The main purpose of this paper is to present bounds of the discrepancy between an infinitely divisible distribution function with finite second moment and the normal as well as the poisson distribution function. Special emphasise is put on explicit numerical constancts involved in the error bounds. To improve these estimates an EDGEWORTH expansion in the smooth case and a POISSON–CHARLEIR expansion in the lattice case are developed. Some applications to the POISSON shot noise illustrate the obtained results

ISSN:0233-1888    

DOI:10.1080/02331889108802342

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

Iterated weighted least squares (IWLS) is investigated for estimating the regression coefficients in a linear model with symmetrically distributed errors. The variances of the errors are not specified; it is neither assumed that they are unknown functions of the explanatory variables nor that they are given in some parametric way IWLS is carried out in a random number of steps, of which the first one is OLS. In each step the error variance at time t is estimated with a weighted sum of m squared residuals in the neighbourhood of t and the coefficients are estimated using WLS. Furthermore an estimate of the covariance matrix is obtained. If this matrix is somehow smaller than the one before, a new step is carried out unless an upper bound has been reached.

ISSN:0233-1888    

DOI:10.1080/02331889108802331

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

In the variance components model, we study the estimation of the parametric functionswe derived the conditions of unbiased estimability, and the expression of locally minimum variance unbiased estimation of parametric function ?

ISSN:0233-1888    

DOI:10.1080/02331889108802332

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

Some useful equiblock-sized and unequireplicated incomplete block designs with ri≤0, k≤0, have been obtained by dualizing equireplicate balanced block designs

ISSN:0233-1888    

DOI:10.1080/02331889108802333

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

In this paper the problem of estimating a power of the parameler under normalized squared error is shown that under the presence of Type 1 censoring a trivial estimator not depending on the sample can be both minimax and admissible. This contrasts with the case of a Type II censoring model where this anomaly does not occur. Some possible treatments of the problem are discussed

ISSN:0233-1888    

DOI:10.1080/02331889108802334

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

O.E. Barndorff-Nielsen, P. Blaesuld, and .s.Eriksen: Decompsition and Invariance of Measures, and Satistical Tranformation Models. Lecure Note in Statistics, No. 58, Springer-Verlag, New York, Berlin and Heidelberg 989, 147 pp., ISBN 0-387-7131-9 and 3-540-97131-9, DM 36,-.

ISSN:0233-1888    

DOI:10.1080/02331889108802335

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

A simple and unified dervation is given for condition ordering results of likelihood ratios. As application and main result we obtain a simple proof of a generalized version of HUZURBAZARS conjecture without any additional moment conditions. HUZURBAZARS conjecture concerns results of the type: Marginal sufficiency implics sufficiency

ISSN:0233-1888    

DOI:10.1080/02331889108802336

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

A non-negative AR(2) process with exponentially distributed white noise is investigated in the paper. It is assumed that the autoregressive parameters are random variables with a vague prior density. They can be esto,ated by their posterior expectations. Explicit formulas for these estimators are derived and their strong consistency is proved. An approximation to the estimators is proposed which is easier for calculation. The results are illustrated in a simulation study

ISSN:0233-1888    

DOI:10.1080/02331889108802338

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

This paper presents a general approach to nonparametric estimation of unknown distribution functions and related characteristics such as cumulative hazard functions. It is based on the notion of portions of statistical data and on the property of discertely separated distributions of statistical data General assumptions are given under which the corresponding generalized maximum likelihood estimators are consistent and their deviations have asymptotically normal distributions, if the number of portions increases to indinity.

ISSN:0233-1888    

DOI:10.1080/02331889108802339

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor

摘要:

Let be the unknown parameter in the drift coefficient of a diffusion process described by a linear homogeneous stochastic differential equation. Bounds for the equivalence of BAYES and Maximum likelihood estimators of the parameter have been obtained in this paper

ISSN:0233-1888    

DOI:10.1080/02331889108802340

出版商:Akademie-Verlag    

年代:1991

数据来源: Taylor