摘要:

Robust statistics are derived for estimating the error variance when a finite population total or a ratio of totals is estimated from a two-stage cluster sample in which the first-stage sampling fraction is small. The analysis is based on prediction (superpopulation) models that allow for very general correlation structure within clusters. Results are obtained for when all cluster sizes are known and for when sizes are known only for sample clusters. Ratio, regression, and mean-of-ratios statistics are in the class of population total estimators considered.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478247

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

For samples of equal size, thepvalues of Bartlett's, Cochran's, and Hartley's tests for the equality of several variances, obtained under the normality assumption, are shown to underestimate the truepvalues when the parent distribution is a normal or aX21scale mixture or when there is within-sample dependence. The proof uses elements of majorization theory. When the sample sizes are not equal, a partial extension is obtained for the two-sample case.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478248

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

Renewal theory and Bayesian decision theory are used to solve a problem related to counting a large number of items by weighing them. Specifically, a batch is to be obtained containing a given number of items by adding items until their total weight reaches a critical value that can depend on the results of a preliminary sample. Furthermore, the optimal sample size for this preliminary sample is to be determined. The distribution of individual weights is assumed to be normal.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478249

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

Outliers are commonplace in data analysis. Time series analysis is no exception. Noting that the effect of outliers on model identification statistics could be serious, this article is concerned with the problem of time series model specification in the presence of outliers. An iterative procedure is proposed to identify the outliers, to remove their effects, and to specify a tentative model for the underlying process. The procedure is essentially based on the iterative estimation procedure of Chang and Tiao (1983) and the extended sample autocorrelation function (ESACF) model identification method of Tsay and Tiao (1984). An example is given. Properties of the proposed procedure are discussed.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478250

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

A theory of overparameterized structural models is presented. In such a model some “redundant” parameters are involved; the parameter vector is not identified, and the information matrix is not nonsingular. The minimum discrepancy function (MDF) test statistic is shown to have an asymptotic chi-squared distribution almost everywhere for a wide class of discrepancy functions. Asymptotic distribution properties of the MDF estimators are investigated. The factor analysis model is discussed as an example.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478251

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

In the regression modelyt= α + βxt+utwhere the disturbancesutmay be autocorrelated andxtfollows an ARIMA(p, 1,q) process, it is shown that the asymptotic distributions of the ordinary and generalized least squares estimates of β are identical.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478252

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

Two new classes of robust estimates for ARMA models are introduced: estimates based on residual autocovariances (RA estimates), and estimates based on truncated residual autocovariances (TRA estimates). A heuristic derivation of the asymptotic normal distribution is given. We also perform a Monte Carlo study to compare the robustness properties of these estimates with the least squares,M, and GM estimates. In this study we consider observations that correspond to a Gaussian model with additive outliers. The Monte Carlo results show that RA and TRA estimates compare favorably with respect to least squares,M, and GM estimates.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478253

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

The distribution of the number of classes observed, when sampling without replacement from a finite population comprised ofkclasses, is derived and factorial moments are given. The mean and up to the fourth central moment are presented. A simple application to an ecological problem illustrates the computations, and other potential applications are discussed.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478254

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

The maximum likelihood estimator of the binomial parameterncan be highly unstable, resulting in huge fluctuations in the estimate when the data are only slightly perturbed. A method of assessing this sensitivity, based on perturbing the log-likelihood function, is proposed and examined.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478255

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

Tsui (1984) showed that many Clevenson—Zidek-type Poisson means estimators are better than the usual estimator under the normalized squared error loss (1.1), even when the underlying distributions are negative binomial. This article shows that this dominance result is still true when the underlying distributions belong to a much larger class of distributions. This class includes all mixtures of Poisson distributions, and hence the negative binomial result is a very special case.

ISSN:0162-1459    

DOI:10.1080/01621459.1986.10478256

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor