The influence of violations of assumptions on multilevel parameter estimates and their standard errors
نویسندگان
چکیده
A crucial problem in the statistical analysis of hierarchically structured data is the dependence of the observations at the lower levels. Multilevel modeling programs account for this dependence and in recent years these programs have been widely accepted. One of the major assumptions of the tests of signi2cance used in the multilevel programs is normality of the error distributions involved. Simulations were used to assess how important this assumption is for the accuracy of multilevel parameter estimates and their standard errors. Simulations varied the number of groups, the group size, and the intraclass correlation, with the second level residual errors following one of three non-normal distributions. In addition asymptotic maximum likelihood standard errors are compared to robust (Huber/White) standard errors. The results show that non-normal residuals at the second level of the model have little or no e8ect on the parameter estimates. For the 2xed parameters, both the maximum likelihood-based standard errors and the robust standard errors are accurate. For the parameters in the random part of the model, the maximum likelihood-based standard errors at the lowest level are accurate, while the robust standard errors are often overcorrected. The standard errors of the variances of the level-two random e8ects are highly inaccurate, although the robust errors do perform better than the maximum likelihood errors. For good accuracy, robust standard errors need at least 100 groups. Thus, using robust standard errors as a diagnostic tool seems to be preferable to simply relying on them to solve the problem. c © 2003 Elsevier B.V. All rights reserved.
منابع مشابه
Assessment and Robust Analysis of Survey Errors
Sharon L. Christ: Assessment and Robust Analysis of Survey Errors (Under the direction of Kenneth Bollen) Random measurement error and errors due to complex sampling designs may have deleterious effects on the quality of parameter estimates. This dissertation is comprised of three research papers that provide 1) an assessment of random measurement error through estimation of reliability using l...
متن کاملLiu Estimates and Influence Analysis in Regression Models with Stochastic Linear Restrictions and AR (1) Errors
In the linear regression models with AR (1) error structure when collinearity exists, stochastic linear restrictions or modifications of biased estimators (including Liu estimators) can be used to reduce the estimated variance of the regression coefficients estimates. In this paper, the combination of the biased Liu estimator and stochastic linear restrictions estimator is considered to overcom...
متن کاملComparison of Maximum Likelihood Estimation and Bayesian with Generalized Gibbs Sampling for Ordinal Regression Analysis of Ovarian Hyperstimulation Syndrome
Background and Objectives: Analysis of ordinal data outcomes could lead to bias estimates and large variance in sparse one. The objective of this study is to compare parameter estimates of an ordinal regression model under maximum likelihood and Bayesian framework with generalized Gibbs sampling. The models were used to analyze ovarian hyperstimulation syndrome data. Methods: This study use...
متن کاملSufficient Sample Sizes for Multilevel Modeling
An important problem in multilevel modeling is what constitutes a sufficient sample size for accurate estimation. In multilevel analysis, the major restriction is often the higher-level sample size. In this paper, a simulation study is used to determine the influence of different sample sizes at the group level on the accuracy of the estimates (regression coefficients and variances) and their s...
متن کاملIgnoring Clustering in Confirmatory Factor Analysis: Some Consequences for Model Fit and Standardized Parameter Estimates.
In many situations, researchers collect multilevel (clustered or nested) data yet analyze the data either ignoring the clustering (disaggregation) or averaging the micro-level units within each cluster and analyzing the aggregated data at the macro level (aggregation). In this study we investigate the effects of ignoring the nested nature of data in confirmatory factor analysis (CFA). The bias ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- Computational Statistics & Data Analysis
دوره 46 شماره
صفحات -
تاریخ انتشار 2004