欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
本文讨论具有ARIMA（0，1，0）对称误差的非线性模型的异方差检验和局部影响分析．对称误差分布族包括正态，t，power exponential，logisticsⅠ，Ⅱ，污染正态等所有对称连续分布．文章首先导出了关于白噪声异方差检验的score统计量及其调整形式，然后对模型进行了局部影响分析，得到了基于似然函数扰动和反应变量扰动的诊断统计量．最后，利用实际数据说明了检验方法的应用，并用Monte Carlo模拟方法研究了异方差检验统计量的检验功效．
英文摘要：
In this paper,we discuss the heteroscedastic test and local influence in nonlinear models with ARIMA （0, 1, 0） symmetrical errors. The symmetrical errors distribution class includes all symmetrical continuous distributions such as normal, Student-t, power exponential,Logistics Ⅰ and Ⅱ ,contaminated normal, so on. First, score test statistic and its adjustment form of the heteroscedastic test are derived. Then,we discuss the local influence analysis of the model and derive diagnostic statistics which based on likelihood functions perturbation and response perturbation. Finally, a real data is used to illustrate the applying of our testing methods, and the properties of test statistics are investigated through Monte Carlo simulations.
同期刊论文项目
同项目期刊论文
Heteroscedasticity and/or autocorrelation diagnostics in nonlinear models with AR(1) and symmetrical
Testing for departures from nominal dispersion in generalized nonlinear models with varying dispersi
期刊信息
位置：