|
1. |
Statistics |
|
Statistics,
Volume 21,
Issue 1,
1990,
Page 1-1
B. Droge,
Preview
|
PDF (167KB)
|
|
ISSN:0233-1888
DOI:10.1080/02331889008802218
出版商:Akademie-Verlag
年代:1990
数据来源: Taylor
|
2. |
Comparing generalized mixed estimators with respect to covariance matrix in a linear regression model |
|
Statistics,
Volume 21,
Issue 1,
1990,
Page 3-8
Erkki P. Liski,
Preview
|
PDF (1697KB)
|
|
摘要:
In this paper THEIL'S mixed estimator (THEIL 1963) is generalized so that all the full rank assumptions of the original definition are removed. Necessary and sufficient conditions are proved for superiority of a generalized mixed estimator over another generalized mixed estimator with respect to covariance matrix for all estimable parametric functions K?. Corresponding generalized results for several important subclasses of mixed estimators, as for restricted least squares estimators, are obtained. It is alos pointed out, that similar kindsl of weaker superiority statements can be proved for a fixed parametrie function K?
ISSN:0233-1888
DOI:10.1080/02331889008802219
出版商:Akademie-Verlag
年代:1990
数据来源: Taylor
|
3. |
Almost exact distributions of estimators I-low dimensional nonlinear regression |
|
Statistics,
Volume 21,
Issue 1,
1990,
Page 9-19
Andrej PÁzman,
Preview
|
PDF (2739KB)
|
|
摘要:
The nonlinear regression model with normally distributed errors is considered. A recently obtained approximation of the density of the maximum likelihood (least squares) estimators is discussed and improvements of the approximation are proposed for small dimensions of the parameter space. The obtained approximations can be considered as ?almost exact? in a sense defined in the paper. These approximations are equivariant, i.e. they behave in the same way as densities when the parameters of the regression model are changed. The approach used in the paper is geometric
ISSN:0233-1888
DOI:10.1080/02331889008802220
出版商:Akademie-Verlag
年代:1990
数据来源: Taylor
|
4. |
Book reivew |
|
Statistics,
Volume 21,
Issue 1,
1990,
Page 20-20
Preview
|
PDF (405KB)
|
|
ISSN:0233-1888
DOI:10.1080/02331889008802221
出版商:Akademie-Verlag
年代:1990
数据来源: Taylor
|
5. |
Almost exact distributions of estimators ii- hat nonlinear regression models |
|
Statistics,
Volume 21,
Issue 1,
1990,
Page 21-33
Andrej PÁzman,
Preview
|
PDF (3252KB)
|
|
摘要:
We consider Gaussian nonlinear regression models with constant information matrix ( = models with constant asymptotic variance) and models which are such after a repararnetrization (= “flat models”), including all one-dimensional nonlinear regression models. In is shown that a recently obtained nonasymptotical approximation of the probability density of the miximum likelihood (= least squares) estimator is particularly good in flat models. It is proved that under this approximative density the gradient of the squared distance between the true and the estimated means of the observed vector is nearly a normal random vector in models with constant information matrix. This allows to construct almost exact confidence regions in flat models, and to obtain approximative moments of the estimators
ISSN:0233-1888
DOI:10.1080/02331889008802222
出版商:Akademie-Verlag
年代:1990
数据来源: Taylor
|
6. |
Book reviews |
|
Statistics,
Volume 21,
Issue 1,
1990,
Page 34-34
W. Winkler,
Preview
|
PDF (265KB)
|
|
摘要:
P. Weiss: Stochastische Modelle fur Anwender. B.G. Teubner, Stuttgart 1978, 192 S.,DM 36,–.
ISSN:0233-1888
DOI:10.1080/02331889008802223
出版商:Akademie-Verlag
年代:1990
数据来源: Taylor
|
7. |
E-optimal 2ksaturated designs |
|
Statistics,
Volume 21,
Issue 1,
1990,
Page 35-44
C Bagiatis,
Preview
|
PDF (2268KB)
|
|
ISSN:0233-1888
DOI:10.1080/02331889008802224
出版商:Akademie-Verlag
年代:1990
数据来源: Taylor
|
8. |
Estimation in random translation models |
|
Statistics,
Volume 21,
Issue 1,
1990,
Page 45-55
Ludger Ruschehdorf,
Preview
|
PDF (3074KB)
|
|
摘要:
The finite sample estimation theory of models with a random location parameter is investigated and some properties of this nonparametric model are compared to corre-sponding properties of the parametric location model. The UMVU-property is very rare in both models and is related to the independence of a maximal ancillary statistic. In the random translation model one has a simple construction method for (local) MVU-estimators and one can justify these estimators by an conditionally principle. In contrast to that in the parametric location models the transition to conditional models does not- simplify the construction and justification. On the other hand the random translation models have larger tangent cones resulting in larger bounds for the estimation of functionals than in the parametric location models. We make some remarks on the asymptotic estimation. problem. The construction of asymptotically efficient estimators remains unsolved
ISSN:0233-1888
DOI:10.1080/02331889008802225
出版商:Akademie-Verlag
年代:1990
数据来源: Taylor
|
9. |
Book reviews |
|
Statistics,
Volume 21,
Issue 1,
1990,
Page 56-56
Silvelyn Zwanzig,
B. Seifert,
Preview
|
PDF (327KB)
|
|
摘要:
H. Scheneeweiss, H.J. Mittag: Lineare Modelle mit fehlerbehatteten Daten. PhysicaVerlag, Heidelberg – Wien 1986, XVIII, 504 S., 50 Abb., 18 Tab.; ISBN 3-7908-0320-0
ISSN:0233-1888
DOI:10.1080/02331889008802226
出版商:Akademie-Verlag
年代:1990
数据来源: Taylor
|
10. |
Location, skewness and tailweight in lssense: a coherent approach |
|
Statistics,
Volume 21,
Issue 1,
1990,
Page 57-74
Jean Averous,
Michel Meste,
Preview
|
PDF (4424KB)
|
|
摘要:
For an univariate distribution F, the usual functional parameters describing location, dispersion and skewness are highly connected with the median and so, may be interpreted as Li tools. In previous papers, the authors defined location functions generalizing M-Location parameters. These functions are used here to extend to Lithe Litools mentionned above. For any s&egs0t a tailweight definition, coherent with the Ls+i sense is introduced. It is shown that the Lsdescription of F can be reduced to the Li description of a transformed distribution FsAsymptotic results are given for s = 1
ISSN:0233-1888
DOI:10.1080/02331889008802227
出版商:Akademie-Verlag
年代:1990
数据来源: Taylor
|
|