首先刻画了非线性随机效应模型的异方差类型，进而研究了非线性随机效应模型的异方差检验；
We first develop the tests for varying dispersion in two special generalized nonlinear models of longitudinal data: (1) logistic nonlinear models in binomial data;
和普通的非线性回归模型一样，具有相关误差的非线性模型也存在异方差检验问题，但通常还要检验相关性。
As in ordinary regression models, the problem of the heteroscedasticity test still exists in nonlinear models with correlated errors, but, the test for correlation also needs to be considered.
应用极值理论，通过极值指数估计量，提出了一种可行的对异方差的检验方法。
One kind of heteroscedasticity testing method was proposed through extreme value theory and extreme value index estimator.
通过ARCHLM检验认为BD I对数序列存在高阶ARCH效应，并用GARCH（1，1）模型消除残差序列的条件异方差性。
High-level ARCH effect is certification in the BDI logarithm process by ARCH LM test, GARCH(1,1)model is used to eliminate the conditional heteroscedasticity.
本文分两节对门限自回归模型中自回归条件异方差的广义谱密度检验进行了讨论。在第一节中，我们介绍了广义谱密度检验。
This thesis is composed of two sections in which we discuss generalized spectral density test of conditional autoregressive heteroscedasticity for threshold autoregressive model.
但实际研究中往往有很多参数不服从假设的分布，针对这一以往方法的缺陷，提出了异方差的游程检验方法。
However, in practice many parameters do not satisfy those hypothesized distribution. In order to handle the defect of normal heteroscedasticity testing means, we pose a new mean, that is runs test.
第二章系统讨论了具有一致相关的纵向数据模型中异方差和相关性的检验问题。
Chapter 2 studies the tests for heteroscedasticity and correlation in longitudinal data model with uniform correlation covariance structure.
文章介绍了异方差模型，研究和分析了异方差的检验和利用加权最小二乘法消除异方差对模型的影响。
They are all used with the hypothesis of homoscedasticity. Heteroscedasticity will danger accuracy of the model. This thesis introduces heteroscedasticity to the reader.
文章介绍了异方差模型，研究和分析了异方差的检验和利用加权最小二乘法消除异方差对模型的影响。
They are all used with the hypothesis of homoscedasticity. Heteroscedasticity will danger accuracy of the model. This thesis introduces heteroscedasticity to the reader.
应用推荐