Dual balance correction in ANOVA repeated measures by missing data


Abstract


The repeated measures ANOVA is being used in study of longitudinal data. In its standard form, the method requires full data. In practice, this is not always possible. In this paper, we consider an approach in which the missing data are neither ignored nor imputed. The   incomplete data result in biased estimations of the time mean. This bias consists of the individual and the group components which are constructed using Markov sequence. The last adjustment is necessary in cases of different numbers of missing values incompared groups.

DOI Code: 10.1285/i20705948v10n1p146

Keywords: ANOVA Repeated Measures; missing data; centralization of model; Markov sequence

References


Afifi A., Azen S. (1972). The Analysis of Variance. Academic Press, New York.

Alexeyeva N. (2013). Analysis of biomedical systems. Reciprocity. Ergodicity. Synonymy. Publishing of the Saint-Petersburg State University, Saint-Petersburg.

Alexeyeva N.P., Tatarinova A.A., Bondarenko B.B. et al. (2011). Analysis of repeated cardiological incomplete data based on ergodic centralization of model. Bulleten of

Almazov Center, 3(8):59–63.

Fitzmaurice G., Davidian M., Verbeke G., Molenberghs G., editor (2008). Longitudinal Data Analysis, chapter 5. Chapman & Hall/CRC Handbooks of Modern Statistical Methods. Chapman and Hall/CRC, Boca Raton.

Hedden S.L., Woolson R.F., Carter R.E. et al. (2009). The impact of loss to follow-up hypothesis tests of the treatment effect for several statistical methods in substance abuse clinical trials. Journal of Substance Abuse Treatment, 37:54–63.

Krupitsky E., Zvartau E., Verbitskaya E., Alexeyeva N. at al. (2012). Double-blind, randomized, placebo-controlled studies of the use of implantable naltrexone forms retard (prodetoksona) for relapse prevention of opium addiction. Addiction issues, 6:3–27.

Myers W.R. (2000). Handling missing data in clinical trials: an overview. Drug Information Journal, 34:525–533.

Rubin D. B. (1976). Inference and missing data. Biometrika, 63:581–592.

Scheffe H. (1999). The Analysis of Variance. John Wiley and Sons, Canada.

Ufliand A., Alexeyeva N. (2014). The dependence of the ergodicity on the time effect in the repeated measures anova with missing data based on the unbiasedness recovery. In Topics in Statistical Simulation, Springer Proceedings in Mathematics and Statistics. 114, pages 517–527. Springer Science + Business Media New York 2014.


Full Text: pdf
کاغذ a4

Creative Commons License
This work is licensed under a Creative Commons Attribuzione - Non commerciale - Non opere derivate 3.0 Italia License.