共查询到20条相似文献,搜索用时 625 毫秒
1.
Bernhard F. Arnold 《Metrika》1996,44(1):119-126
In this paper an approach is presented how to test fuzzily formulated hypotheses with crisp data. The quantitiesα andβ, the probabilities of the errors of type I and of type II, are suitably generalized and the concept of a best test is introduced.
Within the framework of a one-parameter exponential distribution family the search for a best test is considerably reduced.
Furthermore, it is shown under very weak conditions thatα andβ can simultaneously be diminished by increasing the sample size even in the case of testingH
0 against the omnibus alternativeH
1: notH
0, a result completely different from the case of crisp setsH
0 andH
1: notH
0. 相似文献
2.
Summary Dynamic exponential family regression provides a framework for nonlinear regression analysis with time dependent parametersβ
0,β
1, …,β
t, …, dimβ
t=p. In addition to the familiar conditionally Gaussian model, it covers e.g. models for categorical or counted responses. Parameters
can be estimated by extended Kalman filtering and smoothing. In this paper, further algorithms are presented. They are derived
from posterior mode estimation of the whole parameter vector (β′0, …,β′t) by Gauss-Newton resp. Fisher scoring iterations. Factorizing the information matrix into block-bidiagonal matrices, algorithms
can be given in a forward-backward recursive form where only inverses of “small”p×p-matrices occur. Approximate error covariance matrices are obtained by an inversion formula for the information matrix, which
is explicit up top×p-matrices.
Heinz Leo Kaufmann, my friend and coauthor for many years, died in a tragical rock climbing accident in August 1989. This
paper is dedicated to his memory. 相似文献
3.
Prof. Dr. A. Irle 《Metrika》1987,34(1):107-115
Summary LetX
1,X
2, ... form a sequence of martingale differences and denote byZ(a, α) = sup n (S
n
−an
α)+ the largest excess forS
n
=X
1 + ... +X
n
crossing the boundaryan
α. We give a sufficient condition for the finiteness ofEZ(a, α)β which is formulated in terms of bounds forE(X
i
+
p
andE(|X
i
|γ|X
1, ...,X
i-1), whereα, β, γ, p are suitably related. This general result is then applied to the case of independent random variables. 相似文献
4.
Summary A general class of estimators for estimating the population mean of the character under study which make use of auxiliary
information is proposed. Under simple random sampling without replacement (SRSWOR), the expressions of Bias and Mean Square
Error (MSE), up to the first and the second degrees of approximation are derived. General conditions, up to the first order
approximation, are also obtained under which any member of this class performs more efficiently than the mean per unit estimator,
the ratio estimator and the product estimator. The class of estimators in its optimum case, under the first degree approximation,
is discussed. It is shown that it is not possible to obtain optimum values of parameters “a”, “b” and “p”, that are independent of each other. However, the optimum relation among them is given by (b−a)p=ρ C
y/C
x. Under this condition, the expression of MSE of the class is that of the linear regression estimator. 相似文献
5.
Summary A series of 375 “Monte-Carlo” realizations were performed on a number of groupings of 100 samples each fromt distinct binomial populations (t=3, 4, 5, 7, 9) in order to obtain estimates of the power function of theV andX
2 index of dispersion tests in detection of a “trend” in the binomial probabilities. 相似文献
6.
N. K. Jaiswal 《Metrika》1961,4(1):107-125
Summary Time dependent solution of the queuing system characterized by a general independent input, exponential service time distribution
and a finite waiting space, has been first investigated by using the “phase method”. On finding the waiting room full, the
customers then arriving may be turned away or the first customer may wait outside and the input process may be stopped till
the customer then being served, completes its service. Steady state solutions of both these problems have been obtained and
the difference in the operational behaviour of the two systems has been pointed out. For a 2-Erlang arrival distribution,
the queuing parameters have been evaluated for different values ofρ
r
andN. 相似文献
7.
B. M. Bennett 《Metrika》1970,15(1):6-8
Summary LetN=[n
ij
] (i=1, …,r;j=1, …,c) be the matrix of observed frequencies in anr×c contingency table fromr possibly different multinomial populations with respective probabilitiesp
i
=(p
i1, …,p
ic
).Freeman andHalton have proposed an exact conditional test for the hypothesisH
0 :p
i
=(p
1, …p
c
) of the exact test is derived. Numerical values forβ(p) were previously computed for the special case:r=3,c=2 [Bennett andNakamura, 1964]. 相似文献
8.
Summary LetX=(X
ij
)=(X
1, ...,X
n
)’,X’
i
=(X
i1, ...,X
ip
)’,i=1,2, ...,n be a matrix having a multivariate elliptical distribution depending on a convex functionq with parameters, 0,σ. Let ϱ2=ϱ
2
-2
be the squared multiple correlation coefficient between the first and the remainingp
2+p
3=p−1 components of eachX
i
. We have considered here the problem of testingH
0:ϱ2=0 against the alternativesH
1:ϱ
1
-2
=0, ϱ
2
-2
>0 on the basis ofX andn
1 additional observationsY
1 (n
1×1) on the first component,n
2 observationsY
2(n
2×p
2) on the followingp
2 components andn
3 additional observationsY
3(n
3×p
3) on the lastp
3 components and we have derived here the locally minimax test ofH
0 againstH
1 when ϱ
2
-2
→0 for a givenq. This test, in general, depends on the choice ofq of the familyQ of elliptically symmetrical distributions and it is not optimality robust forQ. 相似文献
9.
Dr. P. N. Rathie 《Metrika》1972,18(1):216-219
Equivalence of the generalized entropyH
β (P, Φ
t
) defined in this paper andKapur’s entropy of orderα and typeβ, ie.H
α
β
(P), is established. The results given recently byCampbell follow as special cases.
International Conference on System Sciences, Honolulu, January 1968. 相似文献
10.
In Flak/Schmid (1993) an outlier test for linear processes was introduced. The test statistic bases on a comparison of each
observation with a one-step predictor. It was assumed that an upper bound for the total number of outlierss
n is known, wheren denotes the sample size. The asymptotic distribution of the test statistic was derived under the assumption thats
n/n → 0 ands
n → ∞ asn → ∞. This note deals with the asymptotic behaviour of this quantity, ifs
n/n →p
0 ∈ (0, 1). 相似文献
11.
A distributionF is said to be “more IFR” than another distributionG ifG
−1
F is convex. WhenF(0) =G(0) = 0, the problem of testingH
0 :F(x) =G (θx) for someθ > 0 andx ⩾ 0, against the alternativeH
A:F is more IFR thanG, is considered in this paper. Both cases, whenG is completely specified (one-sample case) and when it is not specified but a random sample form it is available (two-sample
case) are considered. The proposed tests are based onU-statistics. The asymptotic relative efficiency of the tests are compared with several other tests and the test statistics
remain asymptotically normal under certain dependency assumptions.
Research supported in part by a grant from the US Air Force Office of Scientific Research. 相似文献
12.
Holger Dette 《Metrika》1993,40(1):37-50
The optimal design problem for the estimation of several linear combinationsc′
l
ϑ (l=1, …,m) is considered in the usual linear regression modely=f′(x)ϑ (f(x) ∈ ℝ
k
,ϑ ∈ ℝ
k
). An optimal design minimizes a (weighted)p-norm of the variances of the least squares estimates for the different linear combinationsc′
l
ϑ. A generalized Elfving theorem is used to derive the relation of the new optimality criterion to theE-optimal design problem. It is shown that theE-optimal design for the parameterϑ minimizes such a (weighted)p-norm whenever the vectorc=(c′
1, …, c′k)′ is an inball vector of a symmetric convex and compact “Elfving set” in. 相似文献
13.
Prof. Dr. W. Stute 《Metrika》1992,39(1):257-267
LetX
1, ...,X
n
be an i.i.d. sample from some parametric family {θ :θ (Θ} of densities. In the random censorship model one observesZ
i
=min (X
i
,Y
i
) andδ
i
=1{
x
i
≤Y
i}, whereY
i
is a censoring variable being independent ofX
i
. In this paper we investigate the strong consistency ofθ
n
maximizing the modified likelihood function based on (Z
i
,δ
i
, 1≤i≤n. The main result constitutes an extension of Wald’s theorem for complete data to censored data.
Work partially supported by the “Deutsche Forschungsgemeinschaft”. 相似文献
14.
Prof. Dr. T. Royen 《Metrika》1990,37(1):145-154
Summary It is proved that for any fixed argument the sequence (P
k) of the distribution functions of the ranges ofk i.i.d. univariate random variables is log-concave if the random variables have a log-concave density. If the support of the
distribution is an infinite interval and the density is monotonous then the theorem holds also with “log-convex” instead of
“log-concave”. The resulting inequalities can be used by a quick algorithm for closed maximum range test procedures for all
pairwise comparisons (Royen 1988, 1989a, 1989b). Under the above assumptions the application of this algorithm can be extended
e.g. to pairwise comparisons of variances. 相似文献
15.
Prof. Dr. J. Pfanzagl 《Metrika》1970,15(1):30-39
Summary
Lehmann [p. 83] has shown that some families of probability measures with monotone likelihood ratios (m.l.r.) admit median unbiased
estimates which are optimum in the sense that among all median unbiased estimates they minimize the expected loss for any
loss function which assumes its minimal value zero for the “true” parameter value and is nondecreasing as the parameter moves
away from the true value in either direction. This very strong optimum property was proved under the assumption that all probability
measures of the m.l.r.-family have continuous distribution functions, that they are mutually absolutely continuous and that
each element of the support is the median of somep-measure of the family. This result does therefore not cover important cases such as the binomial families or thePoisson family.
The purpose of the present paper is to show the existence ofrandomized median unbiased estimates with the same optimum property for m.l.r.-families which are closed and connected with respect
to the strong topology. Such families are always dominated. We do, however, neither assume that thep-measures are mutually absolutely continuous nor that the distribution functions are continuous. We remark that the use of
randomized estimates is indispensable here because nonrandomized median unbiased estimates do not always exist in the general
case. 相似文献
16.
J. K. Sengupta 《Metrika》1970,15(1):59-70
Summary The problem of statistical distribution of the optimal objective function under the so-called active approach of stochastic
linear programming is investigated here from two interrelated aspects. First, the active approach is viewed as a method of
decomposition. Second, some results on the asymptotic form of distribution of extreme values are utilized to derive the asymptotic
form of the distribution of the maximand under the active approach.
Research done under the partial support of the U.S. National Science Foundation Grant No. 420-04-62 at the Department of Economics,
Iowa State University. Some of the work related to this paper may be found in the following references:Sengupta, J. K., G. Tintner, andC. Millham: “On Some Theorems of Stochastic Linear Programming”. Management Science, Vol. 10, October 1963, pp. 143–159.Sengupta, J. K.: “The stability of truncated solutions of stochastic linear programming”. Econometrica, Vol. 34, January 1966. pp. 77–104.Sengupta, J. K.: “On the stability of solution under recursive programming”. Metrika 1966.Sengupta, J. K. andT. Kumar: “An application of sensitivity analysis to a linear programming problem”. Unternehmensforschung, Vol. 9, 1965. 相似文献
17.
Daniel J. Nordman 《Metrika》2008,68(3):351-363
Properties of a “blockwise”empirical likelihood for spatial regression with non-stochastic regressors are investigated for
spatial data on a lattice. The method enables nonparametric confidence regions for spatial trend parameters to be calibrated,
even though non-random regressors introduce non-stationary forms of spatial dependence into the “blockwise” construction. Additionally, the regression results are valid in a general
framework allowing for a variety of behavior in regressor variables as well as the underlying spatial error process. The same
regression method also applies when the regressors are stochastic. 相似文献
18.
Some notions ofL
p
(μ)-completeness resp. totally L
p
(μ)-completeness (1≦p≦∞) are characterized for families of probability distributions dominated by aσ-finite measureμ and their conservation with respect to direct products is proved. Furthermore, it is shown that totallyL
∞(μ)-completeness does not implyL
1(μ)-completeness and that there are families of probability distributions in the i.i.d. case induced by the order statistic,
which are L1(μ)-complete but not totallyL
∞(μ)-complete. 相似文献
19.
Norbert Henze 《Metrika》1997,45(1):121-130
Smooth goodness of fit tests were introduced by Neyman (1937). They can be regarded as a compromise between globally consistent
(“omnibus”) tests of fit and procedures having high power in the direction of a specific alternative. It is commonly believed
that components of smooth tests like, e.g., skewness and kurtosis measures in the context of testing for normality, have special
diagnostic properties in case of rejection of a hypothesisH
0 in the sense that they constitute direct measures of the kind of departure fromH
0. Recent years, however, have witnessed a complete change of attitude towards the diagnostic capabilities of skewness and
kurtosis measures in connection with normality testing. In this paper, we argue that any component of any smooth test of fit
is strictly non-diagnostic when used conventionally. However, a proper rescaling of components does indeed achieve the desired
“directed diagnosis”. 相似文献
20.
Let {v
n(θ)} be a sequence of statistics such that whenθ =θ
0,v
n(θ
0)
N
p(0,Σ), whereΣ is of rankp andθ εR
d. Suppose that underθ =θ
0, {Σ
n} is a sequence of consistent estimators ofΣ. Wald (1943) shows thatv
n
T
(θ
0)Σ
n
−1
v
n(θ
0)
x
2(p). It often happens thatv
n(θ
0)
N
p(0,Σ) holds butΣ is singular. Moore (1977) states that under certain assumptionsv
n
T
(θ
0)Σ
n
−
v
n(θ
0)
x
2(k), wherek = rank (Σ) andΣ
n
−
is a generalized inverse ofΣ
n. However, Moore’s result as stated is incorrect. It needs the additional assumption that rank (Σ
n) =k forn sufficiently large. In this article, we show that Moore’s result (as corrected) holds under somewhat different, but easier
to verify, assumptions.
Research partly supported by the U.S. Army Research Office through the Mathematical Sciences Institute at Cornell University. 相似文献