共查询到20条相似文献,搜索用时 46 毫秒
1.
《统计学通讯:理论与方法》2012,41(24):6038-6053
AbstractIn this paper, the problem of obtaining efficient block designs for incomplete factorial treatment structure with two factors excluding one treatment combination for estimation of dual versus single treatment contrasts is considered. The designs have been obtained using the A-optimal completely randomized designs and modified strongest treatment interchange algorithm. A catalog of efficient block designs has been prepared for m1?=?3, 4 and m2?=?2, b?≤?10 and k?≤?9 and for m1?=?3,4 and m2?=?3, 4, b?≤?10 and k?≤?10. 相似文献
2.
3.
《统计学通讯:理论与方法》2013,42(12):2637-2653
4.
One of the basic parameters in survival analysis is the mean residual life M
0. For right censored observation, the usual empirical likelihood based log-likelihood ratio leads to a scaled c12{\chi_1^2} limit distribution and estimating the scaled parameter leads to lower coverage of the corresponding confidence interval.
To solve the problem, we present a log-likelihood ratio l(M
0) by methods of Murphy and van der Vaart (Ann Stat 1471–1509, 1997). The limit distribution of l(M
0) is the standard c12{\chi_1^2} distribution. Based on the limit distribution of l(M
0), the corresponding confidence interval of M
0 is constructed. Since the proof of the limit distribution does not offer a computational method for the maximization of the
log-likelihood ratio, an EM algorithm is proposed. Simulation studies support the theoretical result. 相似文献
5.
6.
Consider the regression model Yi= g(xi) + ei, i = 1,…, n, where g is an unknown function defined on [0, 1], 0 = x0 < x1 < … < xn≤ 1 are chosen so that max1≤i≤n(xi-xi- 1) = 0(n-1), and where {ei} are i.i.d. with Ee1= 0 and Var e1 - s?2. In a previous paper, Cheng & Lin (1979) study three estimators of g, namely, g1n of Cheng & Lin (1979), g2n of Clark (1977), and g3n of Priestley & Chao (1972). Consistency results are established and rates of strong uniform convergence are obtained. In the current investigation the limiting distribution of &in, i = 1, 2, 3, and that of the isotonic estimator g**n are considered. 相似文献
7.
8.
Amrita Biswas 《统计学通讯:理论与方法》2013,42(5):920-933
The author considers the problem of finding designs insensitive to the presence of an outlier in a treatment-control block design setup for estimating the set of elementary contrasts between the effects of each test treatment and a control treatment. The criterion of robustness suggested by Mandal (1989) in block design setup for estimating a full set of orthonormal treatment contrasts is adapted. A new class viz. partially balanced treatment incomplete block designs (PBTIBD) is introduced and it is shown that balanced treatment incomplete block designs (BTIBD) and PBTIB designs, under certain conditions, are robust in the previous sense. Such designs are important in the sense that the inference on the treatment contrasts under consideration remain unaffected by the presence of an outlier. 相似文献
9.
Bernard Clement 《Revue canadienne de statistique》1975,3(2):203-222
The problem of estimating the effects in a balanced two-way classification with interaction \documentclass{article}\pagestyle{empty}\begin{document}$i = 1, \ldots ,I;j = 1, \ldots ,J;k = 1, \ldots ,K$\end{document} using a random effect model is considered from a Bayesian view point. Posterior distributions of ri, cj and tij are obtained under the assumptions that ri, cj, tij and eijk are all independently drawn from normal distributions with zero meansand variances \documentclass{article}\pagestyle{empty}\begin{document}$\sigma _r^2 ,\sigma _c^2 ,\sigma _t^2 ,\sigma _e^2$\end{document} respectively. A non informative reference prior is adopted for \documentclass{article}\pagestyle{empty}\begin{document}$\mu ,\sigma _r^2 ,\sigma _c^2 ,\sigma _t^2 ,\sigma _e^2$\end{document}. Various features of thisposterior distribution are obtained. The same features of the psoterior distribution for a fixed effect model are also obtained. A numerical example is given. 相似文献
10.
Sastry G. Pantula 《商业与经济统计学杂志》2013,31(1):63-71
Several test criteria are available for testing the hypothesis that the autoregressive polynomial of an autoregressive moving average process has a single unit root. Schwert (1989), using a Monte Carlo study, investigated the performance of some of the available test criteria. He concluded that the actual levels of the test criteria considered in his study are far from the specified levels when the moving average polynomial also has a root close to 1. This article studies the asymptotic null distribution of the test statistics for testing p = 1 in the model Yt = pY t-1 + e t – 0e t-1 as 0 approaches 1. It is shown that the test statistics differ from one another in their asymptotic properties depending on the rate at which 0 converges to 1. 相似文献
11.
Kai Fun Yu 《Revue canadienne de statistique》1989,17(4):377-389
Suppose it is desired to obtain a large number Ns of items for which individual counting is impractical, but one can demand a batch to weigh at least w units so that the number of items N in the batch may be close to the desired number Ns. If the items have mean weight ωTH, it is reasonable to have w equal to ωTHNs when ωTH is known. When ωTH is unknown, one can take a sample of size n, not bigger than Ns, estimate ωTH by a good estimator ωn, and set w equal to ωnNs. Let Rn = Kp2N2s/n + Ksn be a measure of loss, where Ke and Ks are the coefficients representing the cost of the error in estimation and the cost of the sampling respectively, and p is the coefficient of variation for the weight of the items. If one determines the sample size to be the integer closest to pCNs when p is known, where C is (Ke/Ks)1/2, then Rn will be minimized. If p is unknown, a simple sequential procedure is proposed for which the average sample number is shown to be asymptotically equal to the optimal fixed sample size. When the weights are assumed to have a gamma distribution given ω and ω has a prior inverted gamma distribution, the optimal sample size can be found to be the nonnegative integer closest to pCNs + p2A(pC – 1), where A is a known constant given in the prior distribution. 相似文献
12.
《Journal of statistical planning and inference》2001,97(2):385-398
The class of balanced treatment incomplete block designs is generalized to allow for comparison of v1 test treatments and v2 control treatments. The generalized class is equivalent to the class of balanced bipartite block designs considered by Jaggi, Gupta, and Parsad. Some results on design construction and A-optimality are given for small values of v1 and v2. Algorithms are developed for computing simultaneous confidence bounds for all test treatment versus control contrasts. 相似文献
13.
Suppose there are k
1 (k
1 ≥ 1) test treatments that we wish to compare with k
2 (k
2 ≥ 1) control treatments. Assume that the observations from the ith test treatment and the jth control treatment follow a two-parameter exponential distribution and , where θ is a common scale parameter and and are the location parameters of the ith test and the jth control treatment, respectively, i = 1, . . . ,k
1; j = 1, . . . ,k
2. In this paper, simultaneous one-sided and two-sided confidence intervals are proposed for all k
1
k
2 differences between the test treatment location and control treatment location parameters, namely , and the required critical points are provided. Discussions of multiple comparisons of all test treatments with the best
control treatment and an optimal sample size allocation are given. Finally, it is shown that the critical points obtained
can be used to construct simultaneous confidence intervals for Pareto distribution location parameters. 相似文献
14.
In drug development, non‐inferiority tests are often employed to determine the difference between two independent binomial proportions. Many test statistics for non‐inferiority are based on the frequentist framework. However, research on non‐inferiority in the Bayesian framework is limited. In this paper, we suggest a new Bayesian index τ = P(π1 > π2 ? Δ0 | X1,X2), where X1 and X2 denote binomial random variables for trials n1 and n2, and parameters π1 and π2, respectively, and the non‐inferiority margin is Δ0 > 0. We show two calculation methods for τ, an approximate method that uses normal approximation and an exact method that uses an exact posterior PDF. We compare the approximate probability with the exact probability for τ. Finally, we present the results of actual clinical trials to show the utility of index τ. Copyright © 2013 John Wiley & Sons, Ltd. 相似文献
15.
L2Boosting is an effective method for constructing model. In the case of high-dimensional setting, Bühlmann and Yu (2003) proposed the componentwise L2Boosting, but componentwise L2Boosting can only fit a special limited model. In this paper, by combining a boosting and sufficient dimension reduction method, e.g., sliced inverse regression (SIR), we propose a new method for regression, called dimension reduction boosting (DRBoosting). Compared with L2Boosting, the computation of DRBoosting is less intensive and its prediction is better, especially for high-dimensional data. Simulations confirm the advantage of the new method. 相似文献
16.
17.
The basic model in this paper is an AR(1) model with a structural break in the AR parameter β at an unknown time k0. That is, yt = β1yt ? 1I{t ? k0} + β2yt ? 1I{t > k0} + ?t, t = 1, 2, ???, T, where I{ · } denotes the indicator function. Suppose |β1| < 1, |β2| < 1, and {?t, t ? 1} is a sequence of i.i.d. random variables which are in the domain of attraction of the normal law with zero mean and possibly infinite variance, then the limiting distributions for the least squares estimators of β1 and β2 are studied in the present paper, which extend some results in Chong (2001). 相似文献
18.
19.
Jing-Jing Zhang 《统计学通讯:理论与方法》2013,42(13):3993-4017
ABSTRACTConsider the heteroscedastic partially linear errors-in-variables (EV) model yi = xiβ + g(ti) + εi, ξi = xi + μi (1 ? i ? n), where εi = σiei are random errors with mean zero, σ2i = f(ui), (xi, ti, ui) are non random design points, xi are observed with measurement errors μi. When f( · ) is known, we derive the Berry–Esseen type bounds for estimators of β and g( · ) under {ei,?1 ? i ? n} is a sequence of stationary α-mixing random variables, when f( · ) is unknown, the Berry–Esseen type bounds for estimators of β, g( · ), and f( · ) are discussed under independent errors. 相似文献
20.
This article introduces a novel method, named JC 1, for obtaining G-efficient mixture design to fit quadratic models. The advantage of JC 1 method over existing algorithms is that it gives G-efficient designs without need of generating all the extreme vertices, edge centroids and constraint plane centroids of the mixture experimental region. The performance of the new method is illustrated and its comparison is given with popularly used algorithms—Snee (1975) algorithm and Welch (1985) ACED algorithm for second-order (quadratic model) designs and it is observed that JC 1 method performs as well as the existing methods or sometimes better than those with additional advantage of large savings in computational efforts. 相似文献