5. Three Likelihood-Based Methods for Mean and Covariance Structure Analysis with Nonnormal Missing Data
Tóm tắt
Survey and longitudinal studies in the social and behavioral sciences generally contain missing data. Mean and covariance structure models play an important role in analyzing such data. Two promising methods for dealing with missing data are a direct maximum-likelihood and a two-stage approach based on the unstructured mean and covariance estimates obtained by the EM-algorithm. Typical assumptions under these two methods are ignorable nonresponse and normality of data. However, data sets in social and behavioral sciences are seldom normal, and experience with these procedures indicates that normal theory based methods for nonnormal data very often lead to incorrect model evaluations. By dropping the normal distribution assumption, we develop more accurate procedures for model inference. Based on the theory of generalized estimating equations, a way to obtain consistent standard errors of the two-stage estimates is given. The asymptotic efficiencies of different estimators are compared under various assumptions. We also propose a minimum chi-square approach and show that the estimator obtained by this approach is asymptotically at least as efficient as the two likelihood-based estimators for either normal or nonnormal data. The major contribution of this paper is that for each estimator, we give a test statistic whose asymptotic distribution is chisquare as long as the underlying sampling distribution enjoys finite fourth-order moments. We also give a characterization for each of the two likelihood ratio test statistics when the underlying distribution is nonnormal. Modifications to the likelihood ratio statistics are also given. Our working assumption is that the missing data mechanism is missing completely at random. Examples and Monte Carlo studies indicate that, for commonly encountered nonnormal distributions, the procedures developed in this paper are quite reliable even for samples with missing data that are missing at random.
Từ khóa
Tài liệu tham khảo
Arbuckle James L., 1996, Advanced Structural Equation Modeling: Issues and Techniques, 243
Bentler Peter M., 1995, EQS Structural Equations Program Manual
Bentler Peter M., EQS 6 Structural Equations Program Manual
Bentler Peter M., 1985, Multivariate Analysis VI, 9
Bollen Kenneth A., 1993, Testing Structural Equation Models, 111
Dijkstra Theo K., 1981, Latent Variables in Linear Stochastic Models: Reflections on “Maximum Likelihood” and “Partial Least Squares” Methods
Jöreskog Karl G., 1993, LISREL 8 User's Reference Guide
Kline Rex B., 1998, Principles and Practice of Structural Equation Modeling
Lange Kenneth L., 1989, Journal of the American Statistical Association, 84, 881
Little Roderick J.A., 1987, Statistical Analysis with Missing Data
Mardia Kanti V., 1979, Multivariate Analysis
Meng Xiao-Li, Pedlow Steven. 1992. “EM: A Bibliographic Review with Missing Articles.” Pp. 24–27 in Statistical Computing Section, Proceedings of the American Statistical Association.
Muthén Linda, 1998, Mplus User's Guide
Neale Michael C., 1994, “Mx: Statistical Modeling,”, 2
Rovine Michael J., 1994, Latent Variables Analysis: Applications for Developmental Research, 181
Satorra Albert, Bentler Peter M. 1988. “Scaling Corrections for Chi-Square Statistics in Covariance Structure Analysis.” Pp. 308–13 in American Statistical Association 1988 Proceedings of Business and Economics Sections. Alexandria, VA: American Statistical Association.
Satorra Albert, 1991, Applied Stochastic Models and Data Analysis, 555
Satorra Albert, 1994, Latent Variables Analysis: Applications for Developmental Research, 399
Schoenberg Ronald, 1989, LINCS: Linear Covariance Structure Analysis. User's Guide
Shapiro Alexander, 1983, South African Statistical Journal, 17, 33
Shapiro Alexander, 1987, South African Statistical Journal, 21, 39
Yuan Ke-Hai, 1999, Statistica Sinica, 9, 831
Yung Yiu-Fai, 1996, Advanced Structural Equation Modeling Techniques, 195