Abstract
Longitudinal study has become one of the most commonly adopted designs in medical research. The generalized estimating equations (GEE) method and/or mixed effects models are employed very often in causal inferences. The related model diagnostic procedures are not yet fully formalized, and perhaps never will be. The potential causes of major problems are the high variety of the dependence within subjects and/or the number of repeated measurements. A single testing procedure, e.g., run test, is not possible to resolve all model diagnostics problems in longitudinal data analysis. Multiple quantitative indexes for model diagnostics are needed to take into account this variety. We propose eight testing procedures for randomness accompanied with some conventional and/or non-conventional plots to remedy model diagnostics in longitudinal data analysis. The proposed issue in this paper is well illustrated with four clinical studies in Taiwan.
Similar content being viewed by others
References
Basson BR, Kinon BJ, Taylor CC, et al. (2001) Factors influencing acute weight change in patients with schizophrenia treated with olanzapine, haloperidol, or risperidone. J Clinical Psychiatry 62:231–238
Chang YC (2000), Residuals analysis of the generalized linear models for longitudinal data. Stat Med 19:1277–1293
Diggle PJ, Heagerty P, Liang KY, Zeger SL (2002) Analysis of longitudinal data, 2nd edition. Oxford University Press, New York
Hardin JW, Hilbe JM (2003) Generalized estimating equations. Chapman & Hall/CRC, CRC Press LLC, pp 142–158
Honigfeld G, Klett G (1965) The nurses’ observation scale for inpatients evaluation: a new scale for measuring improvement in chronic schizophrenia. J Clin Psychol 21: 65–71
Kamat AR (1958) Contributions to the theory of statistics based on the first and second successive differences. Metron 19: 97–119
Kamat AR, Sathe YS (1962) Asymptotic power of certain test criteria (based on first and second differences) for serial correlation between successive observations. Ann Math Stat 33:186–200
Kay SR, Fiszbein A, Opler LA (1987) The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull 13:261–276
Lane HY, Chang YC, Chiu CC, Chen TT, Lee SH, Chang WH (2002) Influences of patient-related variables on risperidone efficacy for acutely exacerbated schizophrenia: analyses with rigorous statistics. J Clin Psychopharmacol 22:353–358
Lane HY, Chang YC, Cheng YC, Liu GC, Lin XR, Chang WH (2003) Effects of patient demographics, risperidone dosage, and clinical outcome on body weight in acutely exacerbated schizophrenia patients. J Clin Psychiatry 64:316–320
Liang KY, Zeger SL (1986) Longitudinal data analysis using generalized linear models. Biometrika 73:13–22
McCullagh P, Nelder JA (1989) Generalized linear models 2nd ed. Chapman & Hall, New York
O’Brien PC (1976) A test for randomness. Biometrics 32:391–401
Pan DHC, Guo WY, Chang YC, Chung WY, Shiau CY, Wang LW, Wu SM (1998) The effectiveness and factors related to treatment results of gamma knife radiosurgery for meningiomas. Stereotactic Funct Neurosurgery 70 (suppl 1):19–32
SAS Institute Inc. (1999) SAS OnlineDoc(R), Version 8. SAS Institute Inc, Cary, NC
Zeger SL, Liang KY (1986) Longitudinal data analysis for discrete and continuous outcomes. Biometrics 42:121–130
Acknowledgements
The authors wish to thank Associate Editor and referees for their helpful comments and suggestions. We also would like to thank Dr Ko, Mei-Lan for kindly providing the diabetic retinopathy data, the GK study group in VGH-Taipei for providing the meningiomas data, and Dr. Lane, Hsien-Yuan. for providing the schizophrenia’s data.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Yang, K.H., Chang, YC. A note on model diagnostics in longitudinal data analysis. Computational Statistics 21, 571–587 (2006). https://doi.org/10.1007/s00180-006-0015-y
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00180-006-0015-y