Sensitivity of MRQAP Tests to Collinearity and Autocorrelation Conditions.

Sensitivity of MRQAP Tests to Collinearity and Autocorrelation Conditions.
复制标题

DOI:
10.1007/s11336-007-9016-1
复制
发表时间:
2007-12
期刊:
影响因子:
3
通讯作者:
Snijders TA
Snijders TA
中科院分区:
心理学4区
文献类型:
--
作者:
Dekker D;Krackhardt D;Snijders TA

文献摘要

参考文献

被引文献

相似文献

多元回归二次分配程序(MRQAP)测试是对n个对象之间以相关方阵组织的数据的多线性回归模型系数的排列测试。这样的数据结构在社交网络研究中是典型的,其中变量指示给定的一组参与者之间的某种类型的关系。我们提出了一种新的排列方法(称为“双半分”,或DSP),它补充了现有的MRQAP测试方法家族。我们评估了包括数字信号处理器在内的五种方法的统计偏差(I类错误率)和统计能力,这些方法跨越了网络自相关、伪装(混乱器效应的大小)和数据中的偏斜的各种条件。通过三种假设的数据分布来探讨这些条件:正态分布、伽马分布和负二项分布。我们发现Freedman-Lane方法和DSP方法对于这些条件的广泛应用是最健壮的。我们还发现,如果测试统计量是关键的,所有五种方法都表现得更好。最后,我们发现了MRQAP检验的有用性的局限性:对于伽马分布和负二项分布,所有检验都在极端偏度和高刺激性的同时退化。
Multiple regression quadratic assignment procedures (MRQAP) tests are permutation tests for multiple linear regression model coefficients for data organized in square matrices of relatedness among n objects. Such a data structure is typical in social network studies, where variables indicate some type of relation between a given set of actors. We present a new permutation method (called “double semi-partialing”, or DSP) that complements the family of extant approaches to MRQAP tests. We assess the statistical bias (type I error rate) and statistical power of the set of five methods, including DSP, across a variety of conditions of network autocorrelation, of spuriousness (size of confounder effect), and of skewness in the data. These conditions are explored across three assumed data distributions: normal, gamma, and negative binomial. We find that the Freedman–Lane method and the DSP method are the most robust against a wide array of these conditions. We also find that all five methods perform better if the test statistic is pivotal. Finally, we find limitations of usefulness for MRQAP tests: All tests degrade under simultaneous conditions of extreme skewness and high spuriousness for gamma and negative binomial distributions.
DOI: 10.1080/00949650008812035
发表时间: 2000-01-01
影响因子: 1.2
作者:
Legendre, P
通讯作者: Legendre, P
DOI: 10.1037/0022-3514.84.2.340
发表时间: 2003-02-01
影响因子: 7.6
作者:
Gibbons, D;Olk, PM
通讯作者: Olk, PM
DOI: 10.1016/0378-8733(88)90004-4
发表时间: 1988-12-01
期刊: SOCIAL NETWORKS
影响因子: 3.1
作者:
KRACKHARDT, D
通讯作者: KRACKHARDT, D
DOI: 10.1890/0012-9615(1999)069
发表时间: 1999-11-01
影响因子: 6.1
作者:
Arft, AM;Walker, MD;Wookey, PA
通讯作者: Wookey, PA
DOI: 10.1080/00949659908811936
发表时间: 1999-01-01
影响因子: 1.2
作者:
Anderson, MJ;Legendre, P
通讯作者: Legendre, P