Increasing the flexibility of mixed models by using fractional polynomials

Main Article Content

Edijane Paredes Garcia
Luzia Aparecida Trinca
https://orcid.org/0000-0003-1106-8505

Abstract

The class of regression models incorporating Fractional Polynomials (FPs), proposed by Royston and colleagues in the 1990’s, has been extensively studied and shown to be fruitful in the presence of nonlinearity between the response variable and continuous covariates. FP functions provide an alternative to higher-order polynomials and splines for dealing with lack-of-fit. Mixed models may also benefit from this class of curves in the presence of non-linearity. The inclusion of FP functions into the structure of
linear mixed models has been previously explored, though for simple layouts, e.g. a single covariate in the random intercept model. This paper proposes a general strategy for model-building and variable selection that takes advantage of the FPs within the framework of linear mixed models. Application of the method
to three data sets from the literature, known for violating the linearity assumption, illustrates that it is possible to solve the problem of lack-of-fit by using fewer terms in the model than the usual approach of fitting higher-order polynomials.

Article Details

How to Cite
Garcia, E. P., & Trinca, L. A. (2022). Increasing the flexibility of mixed models by using fractional polynomials. Brazilian Journal of Biometrics, 40(4), 469–489. https://doi.org/10.28951/bjb.v40i4.619
Section
Articles
Author Biography

Edijane Paredes Garcia, Universidade Federal do Amazonas

Department of Statistics

References

Ambler, G. & Royston, P. Fractional polynomial model selection procedures: investigation of Type I error rate. Journal of Statistical Computation and Simulation 69, 89–108 (2001).

Ambler, G. & Benner, A.mfp: Multivariable Fractional Polynomials. R package version 1.5.2.2 (2022). https://CRAN.R-project.org/package=mfp.

Aregay, M., Shkedy, Z., Molenberghs, G., David, M.-P. & Tibaldi, F. Model-based estimates oflong-term persistence of induced HPV antibodies: a flexible subject-specific approach. Journal of Biopharmaceutical Statistics 23, 1228–1248 (2013).

Bates, D., Mächler, M., Bolker, B. & Walker, S. Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software 67, 1–48 (2015).

Belsley, D. A., Kuh, E. & Welsch, R. E.Regression Diagnostics: Identifying Influential Data and Sources of Collinearity (John Wiley & Sons, New York, 1980).

Box, G. E. P. & Tidwell, P. W. Transformation of the independent variables. Technometrics 4, 531–550 (1962).

Crainiceanu, C. M. & Ruppert, D. Likelihood ratio tests in linear mixed models with one variance component. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 66, 165–185 (2004).

Demidenko, E.Mixed Models: Theory and Applications with R (John Wiley & Sons, Hoboken,New Jersey, 2013).

Drikvandi, R., Verbeke, G., Khodadadi, A. & Nia, V. P. Testing multiple variance components in linear mixed-effects models. Biostatistics 14, 144–159 (2013).

Garcia, E. P. Uso de polinômios fracionários nos modelos mistos. Doutorado em Biometria (Instituto de Biociências/ UNESP, Botucatu, 2019).

Harrison, D. & Rubinfeld, D. L. Hedonic housing prices and the demand for clean air. Journal of Environmental Economics and Management 5, 81–102 (1978).

Lee, O. E. & Braun, T. M. Permutation tests for random effects in linear mixed models. Biometrics 68, 486–493 (2012).

Littell, R. C., Milliken, G. A., Stroup, W. W. & Wolfinger, R. D. SAS System for Mixed Models (SAS Institute Inc, Cary, North Carolina, 1996).

Long, J. & Ryoo, J. Using fractional polynomials to model non-linear trends in longitudinal data. British Journal of Mathematical and Statistical Psychology 63, 177–203 (2010).

Longford, N. T. Random Coefficient Models (Oxford University Press, Oxford, 1993).

Pinheiro, J. C. & Bates, D. M. Mixed-Effects Models in S and S-PLUS (Springer-Verlag, NewYork, 2000).

Pinheiro, J. C., Bates, D. M., Debroy, S., Sarkar, D. & Team, R. C.nlme: Linear and Nonlinear Mixed Effects Models R package version 3.1-137 (2018). https://CRAN.R-project.org/package=nlme.

R Core Team.R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing (Vienna, Austria, 2022). https://www.R-project.org/.

Regier, M. D. & Parker, R. D. Smoothing using fractional polynomials: an alternative to polynomials and splines in applied research. Wiley Interdisciplinary Reviews: Computational Statistics 7, 275–283 (2015).

Rocha, F. M. M. Seleção de estruturas de covariância para dados com medidas repetidas. Mestrado em Estatística (IME/USP, São Paulo, 2004).

Royston, P. & Altman, D. G. Regression using fractional polynomials of continuous covariates: parsimonious parametric modelling. Applied Statistics 43, 429–467 (1994).

Royston, P. & Sauerbrei, W. Improving the robustness of fractional polynomial models by preliminary covariate transformation: A pragmatic approach. Computational Statistics & Data Analysis 51, 4240–4253 (2007).

Royston, P. & Sauerbrei, W. Multivariable model-building: a pragmatic approach to regression anaylsis based on fractional polynomials for modelling continuous variables (John Wiley & Sons, Chichester, England, 2008).

Saurbrei, W., Meier-Hirmer, C., Benner, A. & Royston, P. Multivariable regression model building by using fractional polynomials: Description of SAS, STATA and R programs. Computational Statistics & Data Analysis 50, 3464–3485 (2006).

Singer, J. M., Rocha, F. M. M. & Nobre, J. S. Graphical tools for detecting departures from linear mixed model assumptions and some remedial measures. International Statistical Review 85, 290–324(2017)

Tilling, K., MacDonald-Wallis, C., Lawlor, D. A., Hughes, R. A. & Howe, L. D. Modelling childhood growth using fractional polynomials and linear splines. Annals of Nutrition and Metabolism 65, 129–138 (2014).

Vonesh, E. F. & Carter, R. L. Mixed-effects nonlinear regression for unbalanced repeated measures. Biometrics 48, 1–17 (1992).