摘要:

SummaryThis keynote address at the 7th Australian Statistical Conference (1984) discusses briefly seven modifications of sample design for improving the usefulness and timeliness of surveys relevant to public poky. These are: better estimates for small domains, cumulating rolling samples, more panel surveys, multipurpose designs for periodic samples, split panel designs (SPD) combining panels with nonoverlapping samples, and frequent collections cumulated for less frequent reporting periods.

ISSN:0004-9581    

DOI:10.1111/j.1467-842X.1986.tb00579.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

SummaryThe only parametric model in current use for axial data from a rotationally symmetric bipolar or girdle distribution on the sphere is the Watson distribution. This paper develops methods for evaluating the model as a fit to data using graphical and formal goodness‐of‐fit tests, and tests of discorda

ISSN:0004-9581    

DOI:10.1111/j.1467-842X.1986.tb00580.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

SummaryThe problem of evaluating the sampling variance of a mean when units are subsampled from a multistage sample is discussed. In particular, a specific problem arising in the sampling of wool where information on the identity of subsampled units is not available is dealt with by introducing a simple model to account for the variability in the unobserved numbers of observations within the multistage categories. Estimation aspects of the wool example are also dealt with.

ISSN:0004-9581    

DOI:10.1111/j.1467-842X.1986.tb00581.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

SummaryMay nonidentity error correlation patterns encountered in regression theory allow computationally efficient techniques for modifying the usual procedures for constructing confidence intervals and performingF‐tests so that valid inferences can be drawn. Such techniques are explored in light of the general theme of this not

ISSN:0004-9581    

DOI:10.1111/j.1467-842X.1986.tb00582.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

SummaryConsistency and asymptotic normality of the maximum likelihood estimator of β in the loglinear model E(yi) = eα+βXi, where yiare independent Poisson observations, 1iaan, are proved under conditions which are near necessary and sufficient. The asymptotic distribution of the deviance test for β=β0is shown to be chi‐squared with 1 degree of freedom under the same conditions, and a second order correction to the deviance is derived. The exponential model for censored survival data is also treated by the same m

ISSN:0004-9581    

DOI:10.1111/j.1467-842X.1986.tb00583.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

ISSN:0004-9581    

DOI:10.1111/j.1467-842X.1986.tb00584.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

SummaryThis paper introduces a new class of designs called equi‐neighboured designs. An equineighboured design has the property that every unordered pair of treatments occurs as nearest neighbours equally frequently at every level. These designs are defined in Section 4 and shown to be balanced when neighbouring observations are correlated. Some equineighboured designs are constructed using a complete set of orthogonal Latin squares. Cyclic equineighboured designs are also define

ISSN:0004-9581    

DOI:10.1111/j.1467-842X.1986.tb00585.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

SummaryTwo characterizations of the uniform distribution on a suitable compact space are proved. These characterizations are applied to a number of particular examples of which the most interesting is the following: ifX,YandZare independentn‐vectors whose components are independent and identically distributed within a vector, then the pairwise independence of the product moment correlation coefficients betweenX,YandZimplies that these vectors are normally distribute

ISSN:0004-9581    

DOI:10.1111/j.1467-842X.1986.tb00586.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

SummaryLetX1X2,.be i.i.d. random variables and let Un= (n r)‐1S̀(n,r)h (Xi1,.,Xir,) be aU‐statistic withEUn= v,vunknown. Assume that g(X1) =E[h(X1,.,Xr) ‐v|X1]has a strictly positive variance s̀2. Further, let a be such that φ(a) ‐ φ(‐a) =α for fixed α, 0<α0, and a confidence interval forvof length 2d,of the formIn,d= [Un,‐d, Un+ d].We assume that VarUnis unknown, and hence, no fixed sample size method is available for finding a confidence interval forvof prescribed width 2d and prescribed coverage probability α Turning to a sequential procedure, letIN(d),dbe a sequence of sequential confidence intervals forv.The asymptotic consistency of this procedure, i.e. limd → 0P(v ∈ IN(d),d)=α follows from Sproule (1969). In this paper, the rate at which |P(v ∈ IN(d),d)converges to α is investigated. We obtain that |P(v ∈ IN(d),d)‐ α| = 0 (d1/2‐(1+k)/2(1+m)), d → 0, where K = max {0,4 ‐ m}, under the condition thatE|h(X1, Xr)|m2. This improves and extends recent results o

ISSN:0004-9581    

DOI:10.1111/j.1467-842X.1986.tb00587.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

SummaryIn this paper, exact solution of Wilks' type‐Bintegral equation has been obtained in its most general form as a series of weighted gamma distributions. This general result then gives the distributions of many test statistics in multivariate analysis. In particular the distributions of Wilks' Λ, the sphericity test criterion, and Bartlett's test statistic, are derived in easily computable fo

ISSN:0004-9581    

DOI:10.1111/j.1467-842X.1986.tb00588.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY