Showing papers in "Calcutta Statistical Association Bulletin in 1984"
TL;DR: In this paper, the identity of two univariate distribution functions F1(x) and F9(x), when there is a sample of fixed size from F1 (x), and the observations (Y) from F2 (x) are drawn sequentially, is investigated.
Abstract: The problem considered in this paper is that of testing the identity of two univariate distribution functions F1(x) and F9(x) when there is a sample of fixed size from F1 (x) and the observations (Y) from F2(x) are drawn sequentially. Two score functions 4> 1(u), 0 < u < 1, l= I, 2, which reflect departures from the null hypothesis in specific ways are taken. Writing F~. (x) for a certain uniformly consistent estimate of F1 (x) based on the first sample, sums of the type S,. 01 = £ 4> 1 (Ft. (Y;)) are calculated for successive observations ;-1 on Y. Observation is stopped as soon as S,. ( 11 reaches a pre-fixed upper bound. The hypothesis is rejected or accepted on the basis of the terminal value of Sn <•>. Different large sample procedures based on this approach are formulated and examined.
18 citations
TL;DR: In this paper, variance balanced incomplete block designs have been constructed in unequal block sizes for situations when suitable BIB designs do not exist for a given number of treatments, where the variance is fixed.
Abstract: In this paper variance balanced incomplete block designs have been constructed in unequal block sizes for situations when suitable BIB designs do not exist for a given number of treatments.
10 citations
TL;DR: In this article, nonparametric estimation of the quantiles of a distribution based on a sample from the corresponding length-biased distribution is considered, along with some representations of this estimator in terms of averages of independent random variables, some limiting results are established.
Abstract: Nonparametric estimation of the quantiles of a distribution based on a sample from the corresponding length-biased distribution is considered. Along with some representations of this estimator in terms of averages of independent random variables, some limiting results are established. The case of reduced quantile processes is also treated briefly.
10 citations
TL;DR: A characterizing property of the generalized gamma distribution based on a conditional distribution has been studied in this article, following a similar approach presented by Patil and Seshadri (1964).
Abstract: A characterizing property of the generalized gamma distribution based on a conditional distribution has been studied in this work following a similar approach presented by Patil and Seshadri (1964). As a particular case a characterization of the Weibull distribution has alse been covered.
6 citations
TL;DR: In this article, connections between efficient and inefficient sampling plans are explored via the concept of sufficiency of statistical experiments, and the problem of comparison of two arbitrary sampling plans is also taken up.
Abstract: Connections between the so called efficient and inefficient Bernoulli sampling plans are explored, via the concept of sufficiency of statistical experiments. From this point of view, many inefficient plans turn out to be quite useful. We also take up the problem of comparison of two arbitrary plans and provide a variety of results. The paper ends with a few conjectures on some naturally raised questions, yet to be satisfactorily settled.
5 citations
TL;DR: For a two-parameter Weibull distribution, moment estimators of the parameters have been developed by choosing orders of two moments (allowing fractions) so that the overall relative efficiency of t....
Abstract: For a two-parameter Weibull distribution, moment estimators of the parameters have been developed by choosing orders of two moments (allowing fractions) so that the overall relative efficiency of t...
3 citations
TL;DR: The concept of block designs with nested rows and columns is applied to balanced block designs [Corsten (1962)] and some methods of construction are given along with the analysis as mentioned in this paper, but the analysis is limited to balanced blocks.
Abstract: The concept of block designs with nested rows and columns is applied to Balanced Block designs [Corsten (1962)]. Some methods of construction are given along with the analysis.
3 citations
TL;DR: These designs may be particularly useful for factorial experiments, and are termed partially efficiency balanced block designs, which are considered in this paper.
Abstract: In this paper we consider some methods for constructing a class of connected block designs, termed as partially efficiency balanced block designs. These designs may be particularly useful for factorial experiments.
2 citations
TL;DR: In this paper, it is shown how to construct a pair of orthogonal v 2 by 2v rectangles for all even v for which a set of t pairwise Latin squares of order v exists.
Abstract: It is shown how to construct a pair of orthogonal v⁄2 by 2v rectangles for all even v. Also, it is shown how to construct a set of t pairwise orthogonal v⁄2 by 2v F-rectangles for all even v for which a set of t pairwise orthogonal Latin squares of order v exists. Some situations where these designs are useful in practice are indicated.
2 citations
TL;DR: In this article, the authors derived an optimal estimator of finite population variance within a general class of quadratic estimators for generalised random permutation models and showed that the class of non-negative unbiased polynomial estimators of variance has also been examined.
Abstract: Considering certain generalised random permutation models we have derived optimal estimators of finite population variance within a general class of quadratic estimators. The class of non-negative unbiased polynomial estimators of variance has also been examined and an optimal estimator within that class has been derived under this model.
1 citations
TL;DR: In this paper, it was shown that every DFR survival probability with a finite mean is stochastically strictly dominated by an exponential mixture, under sufficient conditions, under which such is the case, and the main thrust of their findings is that their bounds are tighter for moderate to heavy departures from exponentiality.
Abstract: We obtain some new reliability bounds for the class of life distributions which are exponential mixtures. It is shown that our bounds often improve on the usual DFR upper bound as well as those contained in the recent results of Shaked, Heyde and Leslie, Hall , and Brown. Simple sufficient conditions, under which such is the case, are developed. The main thrust of our findings is that our bounds are tighter for moderate to heavy departures from exponentiality. Among other results, it is shown that every DFR survival probability with a finite mean is stochastically strictly dominated by an exponential mixture. An application of our methods yields a new bound on the tail of the equilibrium distribution of a DFR renewal process, which can be tighter than the corresponding results of Brown.
TL;DR: In this article, the authors developed sequential and two-stage procedures for estimating the difference and selecting the larger of two truncation parameters of two unknown distributions, which form the nonparametric two-sample versions of the results in Mukhopadhyay and Hamdy (1984), and Mukhopaddhyay (1984).
Abstract: This paper develops sequential and two-stage procedures for (i) estimating the difference, and (ii) selecting the larger of two truncation parameters of two unknown distributions. These form the nonparametric two-sample versions of the results in Mukhopadhyay and Hamdy (1984), and Mukhopadhyay (1984). Some of the basic tools for the corresponding one-sample nonparametric problem are developed in Swanepoel vanWyk (1981).
TL;DR: For the single sample problem of testing for location against one-sided alternatives under the nonparametric set-up, the question of utilization of additional information is examined in this article.
Abstract: For the single sample problem of testing for location against one-sided alternatives under the non parametric set-up, the question of utilization of additional information is examined. Considering the case of the sign test it is shown that if suitable bounds to the ratio of the densities at the median and the known quantile are available it is possible to construct a more sensitive test by using a combination of the sign test statistic and the number of exceedances above the known quantile.
TL;DR: In this paper some methods of construction of neighbour designs having k=4, are developed using the method of differences.
Abstract: In this paper some methods of construction of neighbour designs having k=4, are developed using the method of differences.
TL;DR: In this paper, the best mode of sampling in double sampling with two auxiliary characters has been investigated and it has bee shown that generally the sampling procedure in which all the characters ar observed in independent samples provides the most efficient estimator.
Abstract: In the present note the. best mode of sampling in doubl sampling with two auxiliary characters ha s been in vestigated a nd it has bee shown that generally the sampling procedure in which all the characters ar observed in independent samples provides the most efficient estimator.
TL;DR: In this article, new classes of rules are proposed for selecting from a given set of populations, a subset containing the best population, which have the following property: the probability that the selected set contains the best populations is greater than or equal to P*, a preassigned number.
Abstract: In this paper, new classes of rules are proposed for selecting from a given set of populations, a subset containing the best population. The rules have the following property: The probability that the selected set contains the best population is greater than or equal to P*, a preassigned number. The derivation of the rules is based on Schur concave functions. The rules apply to various types of distributions such as normal and gamma.