该方法的检验统计量称为Z统计量。
在原假设下得到检验统计量的极限分布。
Under null hypothesis, the limit distributions of statistics are obtained.
将秩次转换为正态计分，并以此为基础计算检验统计量。
The ranks were transformed into the normal scores by which the test statistics were calculated.
通过模拟试验研究了检验统计量的小样本性质，并进行了实际数据分析。
The small sample properties of the tests are investigated by Monte Carlo simulation experiments and real data is analyzed for illustration.
本文采用LSTR模型和LM检验统计量，对此问题进行了深入细致的探讨。
This paper gives a thorough discussion on the issue using LSTR model and LM test statistics.
并应用Bootstrap自助法来获得检验统计量的渐近p—值或渐近临界值。
We obtain asymptotic P-value or asymptotic critical value of tests...
利用样本分位数构造检验统计量，给出来自于指数分布总体异常数据的一种检测方法。
A method of detection of outliers from the exponential distribution is given by sample fractile to construct test statistics.
本文提出从HNBUE分布类中检验指数分布的方法，证明检验统计量的渐近正态性和检验的相合性。
This paper proposes a test statistics for testing exponential distribution versus HNBUE, and proves the statistics approximate normal distribution and consistence.
给出了正态总体均值和标准差的最大似然估计（M LE），似然比检验统计量及其渐近分布等结果。
The maximum likelihood estimators(MLE) of means and standard deviations, and the asymptotic distribution of likelihood ratio statistic were given.
此外，通过对10个CNN网络进行KS检验统计量的比较，证实该方法比图形方法具有更好的拟合效果。
Furthermore, it has been verified to have higher goodness-of-fit than graphical methods by comparing the ks test statistics for 10 CNN networks.
最新的文献表明这个提出的修正似然比检验统计量在零假设下的渐近分布是比较简单，并且是容易应用的。
It is shown that the asymptotic null distribution of the modified likelihood ratio test proposed is derived and found to be relatively simple and easily applied.
证明估计的强相合性和渐近正态性，给出似然比检验统计量的极限分布，并讨论基于精确分布的检验问题。
The limit distributions of estimators and likelihood ratio test are given, the strong consistency of estimators is also proved.
研究两次重复试验三因子全效应随机模型误差方差的齐性检验问题，推导了检验统计量和检验规则的具体表达式。
According to the homogeneity test for error variance on total-effect random model with three factors in double repeated experiment, the specific formulae of test statistics and test rule are deduced.
本文给出了制定定时截尾寿命试验下指数分布产品失效率的抽样检验方案的一个新方法，检验统计量是总试验时间。
A sampling inspection plan for the failure rate of GBVE distribution with the test statistic of total experimental time based on Type I censoring is presented.
本文给出了制定定时截尾寿命试验下指数分布产品失效率的抽样检验方案的一个新方法，检验统计量是总试验时间。
This paper gives a new statistical method of working out a sampling inspection plan for the failure rate of exponential populations.
我们证明了在原假设和局部对立假设之下随机加权统计量的条件渐近分布与检验统计量在原假设之下的渐近分布相同。
It is shown that, under both the null and the local alternatives, the random weighting statistics have the same asymptotic distribution as the original test statistic under the null hypotheses.
文中着重分析了在函数模型正确的前提下，采用不正确的随机模型对方差一致性检验统计量的影响，并给出了具体的影响公式。
Supposed the functional model is correct, the influence on the statistic of test for uniform variance with incorrect stochastic model is analyzed in detail and its formula is derived.
以总试验时间为检验统计量，在定时截尾寿命试验下研究GBVE分布产品失效率的抽样检验方案，随机模拟结果显示此方法有效。
A sampling inspection plan for the failure rate of GBVE distribution with the test statistic of total experimental time based on Type I censoring is presented.
在平均值为零或平均值为已知的季节时间序列模型中，根据加权对称估计量提出季节单位根的检验统计量，并求出此统计量的极限分布的表达式。
In this article, we propose seasonal unit root test statistics based on the Weighted Symmetric estimator and derive representation for limiting distributions of the statistics.
在平均值为零或平均值为已知的季节时间序列模型中，根据加权对称估计量提出季节单位根的检验统计量，并求出此统计量的极限分布的表达式。
In this article, we propose seasonal unit root test statistics based on the Weighted Symmetric estimator and derive representation for limiting distributions of the statistics.
应用推荐