Simplified Estimation and Testing in Unbalanced Repeated Measures Designs

被引:1
作者
Spiess, Martin [1 ]
Jordan, Pascal [1 ]
Wendt, Mike [2 ]
机构
[1] Univ Hamburg, Hamburg, Germany
[2] Med Sch Hamburg, Hamburg, Germany
关键词
repeated measures design; unbalanced design; generalized estimating equations; bootstrap; Wald test; task switching paradigm; MATRIX CALCULUS; RECONFIGURATION; INTERFERENCE; INFERENCE; MODELS;
D O I
10.1007/s11336-018-9620-2
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
In this paper we propose a simple estimator for unbalanced repeated measures design models where each unit is observed at least once in each cell of the experimental design. The estimator does not require a model of the error covariance structure. Thus, circularity of the error covariance matrix and estimation of correlation parameters and variances are not necessary. Together with a weak assumption about the reason for the varying number of observations, the proposed estimator and its variance estimator are unbiased. As an alternative to confidence intervals based on the normality assumption, a bias-corrected and accelerated bootstrap technique is considered. We also propose the naive percentile bootstrap for Wald-type tests where the standard Wald test may break down when the number of observations is small relative to the number of parameters to be estimated. In a simulation study we illustrate the properties of the estimator and the bootstrap techniques to calculate confidence intervals and conduct hypothesis tests in small and large samples under normality and non-normality of the errors. The results imply that the simple estimator is only slightly less efficient than an estimator that correctly assumes a block structure of the error correlation matrix, a special case of which is an equi-correlation matrix. Application of the estimator and the bootstrap technique is illustrated using data from a task switch experiment based on an experimental within design with 32 cells and 33 participants.
引用
收藏
页码:212 / 235
页数:24
相关论文
共 50 条
[41]   Multilevel Analysis for Repeated Measures Data in Lambs [J].
Akkol, Suna ;
Karakus, Ferda ;
Cengiz, Firat .
JOURNAL OF AGRICULTURAL SCIENCES-TARIM BILIMLERI DERGISI, 2018, 24 (02) :218-226
[42]   Missing repeated measures data in clinical trials [J].
Pugh, Stephanie L. ;
Brown, Paul D. ;
Enserro, Danielle .
NEURO-ONCOLOGY PRACTICE, 2022, 9 (01) :35-42
[43]   On repeated measures analysis with misspecified covariance structure [J].
Crowder, M .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 2001, 63 :55-62
[44]   Calculating a sample size for a study with repeated measures [J].
Nordgren, Rachel .
JOURNAL OF MOLECULAR AND CELLULAR CARDIOLOGY, 2019, 133 :217-219
[45]   A note on repeated measures analysis for functional data [J].
Smaga, Lukasz .
ASTA-ADVANCES IN STATISTICAL ANALYSIS, 2020, 104 (01) :117-139
[46]   On testing variance components in unbalanced mixed linear model [J].
Širková L. ;
Witkovský V. .
Applications of Mathematics, 2001, 46 (3) :191-213
[47]   Cross-Lagged Mendelian Randomization: Extending Causal Inference in Observational Studies by Integrating Genetic Instrumental Variables with Repeated-Measures Designs [J].
Singh, Madhurbain ;
Verhulst, Brad ;
Vinh, Philip ;
Zhou, Yi ;
Castro-de-Araujo, Luis F. S. ;
Dolan, Conor V. ;
Neale, Michael C. .
BEHAVIOR GENETICS, 2022, 52 (06) :391-391
[48]   Estimation of the limit of detection using information theory measures [J].
Fonollosa, Jordi ;
Vergara, Alexander ;
Huerta, Ramon ;
Marco, Santiago .
ANALYTICA CHIMICA ACTA, 2014, 810 :1-9
[49]   Deductive Semiparametric Estimation in Double-Sampling Designs with Application to PEPFAR [J].
Qian, Tianchen ;
Frangakis, Constantine ;
Yiannoutsos, Constantin .
STATISTICS IN BIOSCIENCES, 2020, 12 (03) :417-445
[50]   Bias correction for estimation of performance measures of a Markovian queue [J].
Almeida, M. A. C. ;
Cruz, F. R. B. ;
Oliveira, F. L. P. ;
de Souza, G. .
OPERATIONAL RESEARCH, 2020, 20 (02) :943-958