日期精度：到日 2274 篇 · 到月 7743 篇 · 到年 0 篇
Ⅰ型删失抽样下双参数指数分布的统计推断
Inferences concerning the location and scale parameters of an exponential distribution are discussed for samples censored at a fixed value. The procedures considered are uniformly most powerful unbiased if the tests are performed with replacement and they seem to be quite reasonable when testing without replacement.
严格递增回归函数的估计
The maximum likelihood estimate of a nondecreasing regression function with independent, normal errors has been studied in detail in the literature. However, it is often constant over one or more intervals and for this reason may be unacceptable if the regression function is believed to be strictly increasing. Using simulation techniques, the least squares line, the maximum likelihood estimate, and a weighted average of these two are compared here. In each case considered,…
最小绝对误差回归的渐近理论
In the general linear model with independent and identically distributed errors and distribution function F, the estimator which minimizes the sum of absolute residuals is demonstrated to be consistent and asymptotically Gaussian with covariance matrix ω2 Q -1, where Q = lim T -1 X'X and ω2 is the asymptotic variance of the ordinary sample median from samples with distribution F. Thus the least absolute error estimator has strictly smaller asymptotic confidence ellipsoids…
选取包含最佳正态总体子集的新程序
A new procedure, called the fiducial procedure, is defined for selecting a subset of a collection of normal populations with common known variance. In the presence of an “indifference zone,” this procedure has probability at least P* of selecting the “best” population. An upper bound is obtained for the smallest indifference zone d* that satisfies the P* condition. Comparisons of relative merit are made between the fiducial procedure, Gupta's (1965) procedure, and an…
若干角度—线性联合分布及相关回归模型
Parametric models are proposed for the joint distribution of bivariate random variables when one variable is directional and one is scalar. These distributions are developed on the basis of the maximum entropy principle and by the specification of the marginal distributions. The properties of these distributions and the statistical analysis of regression models based on these distributions are explored. One model is extended to several variables in a form that justifies the…
估计泊松过程到达率的贝叶斯序贯程序
Let Wn, n = 0, 1, …, be the time until the nth arrival of a Poisson process with rate θ. Using loss L (θ, ) = θ−2(θ − )2 and sampling costs involving cost per arrival and cost per unit time, the Bayes sequential procedure is derived. The large-sample properties of the procedure are then studied in the classical framework, and the optimal stopping time N* is shown to be asymptotically equivalent to n*, that function of θ which minimizes the expected total cost given θ.…
瓮模型在序贯对照临床试验设计中的应用
In the comparison of K treatments, assume that patients appear singly and must be treated immediately. Suppose that patients having the same combination of prognostic factor levels are grouped into the same stratum. If the number of different strata is small, we treat each stratum as a separate independent subtrial. A treatment assignment rule is proposed that forces a small subtrial to be balanced, but tends toward the complete randomization scheme as the size of the…
Box–Cox 变换小样本性质的蒙特卡洛研究
The small-sample properties of models which transform both the dependent and independent variables using the same Box-Cox transformation are investigated in sampling experiments. Bias does not appear to be a serious problem; however, the sign and size of the transformation parameter, which changes the coefficient of variation of the dependent variable, seriously affects the variances of the estimators. Hypothesis tests reveal that t statistics frequently lead to incorrect…
多条回归线平行性秩得分检验的小样本比较
The small-sample behavior of rank score tests due to Adichie and to Sen is compared to the classical parametric F test for parallelism of regression lines. Empirical power functions generated by a series of Monte Carlo simulations reveal that the nonparametric tests are consistently conservative relative to the classical test and suggest that, within the confines of the experimental distributions, the parametric test is the preferred small-sample test for parallelism.
线性判别分类后一组观测下的均值估计
Suppose that a group of observations is known to have come from either one of two normal populations with unknown means and common variance. After the group is classified by the linear discriminant function, it is pooled with the sample from the population to which it belongs to estimate the unknown mean. Weighted estimators of the population means are defined. The biases and mean squared errors of the estimators in the univariate case are obtained. Estimators of the…
估计泊松过程到达率的贝叶斯序贯程序
Let Wn, n = 0, 1, …, be the time until the nth arrival of a Poisson process with rate θ. Using loss L (θ, ) = θ−2(θ − )2 and sampling costs involving cost per arrival and cost per unit time, the Bayes sequential procedure is derived. The large-sample properties of the procedure are then studied in the classical framework, and the optimal stopping time N* is shown to be asymptotically equivalent to n*, that function of θ which minimizes the expected total cost given θ.…
数据收集前聚合问题的决策论方法
We present a decision-theoretic approach to finding optimal aggregation schemes when the aggregation is made at the pre-data-collection stage. A criterion is presented for the optimal aggregation of variables, its properties are analyzed, and it is compared with other criteria in the literature for good aggregation. In the course of determining this criterion, we derive endogenously the disaggregation procedure by which inferences about disaggregated data are made from…
百分位回归：参数化方法
For situations in which no assumptions are explicitly made about error distributions, other model assumptions may imply constraints on the form of the residual distribution. This is true of regression models which assume that percentiles are functions of the regressor variable. Hogg (1975) used a procedure based on signs of residuals to fit linear percentile models. Advantages of making (and testing!) explicit parametric assumptions in this context are illustrated using a…
选择包含最优正态总体的子集的新程序
A new procedure, called the fiducial procedure, is defined for selecting a subset of a collection of normal populations with common known variance. In the presence of an “indifference zone,” this procedure has probability at least P* of selecting the “best” population. An upper bound is obtained for the smallest indifference zone d* that satisfies the P* condition. Comparisons of relative merit are made between the fiducial procedure, Gupta's (1965) procedure, and an…
回归选择策略与揭示的先验
The computation and selection of constrained regressions may be motivated by prior information and, if so, a regression selection strategy reveals the implicit prior. The selection strategies of principal component regression, stepwise regression, and imposing equality constraints are connected with prior densities which are uniform on spheres, hyperbolas, and cones, respectively. Omitting variables in a predetermined order reveals lexicographic priors.
线性不等式约束下线性模型的贝叶斯分析
This article considers the general linear model when the parameter space is subject to linear inequality constraints. A Bayesian analysis of this model is presented using a natural conjugate prior of the mixed type. Expressions are given for the probability that constraints are binding and for the distribution of the parameters. When prior information about the parameters is vague, the Bayesian and sampling methods of model selection are compared. The techniques are applied…
一种稳健的离散程度检验
The simplicity and surprisingly good small-sample power properties of the ordinary sign test motivate its use, in suitably adapted form, as a dispersion test. The proposed test statistic is based on the number of observations lying more than a specified distance from the sample mean. For symmetric and moderately skewed distributions, the proposed test is robust and has good power properties.