1. On the estimation of variance parameters in non-standard generalised linear mixed models: application to penalised smoothing
- Author
-
Paul H. C. Eilers, María Xosé Rodríguez-Álvarez, Dae-Jin Lee, María Durbán, Ministerio de Economía y Competitividad (España), and Epidemiology
- Subjects
Statistics and Probability ,Mixed model ,FOS: Computer and information sciences ,mixed models ,Splines ,Reml ,Restricted maximum likelihood ,Degrees of freedom (statistics) ,Generalised Linear Mixed Models ,Generalised Additive Models ,010103 numerical & computational mathematics ,Estadística ,01 natural sciences ,Generalized linear mixed model ,Theoretical Computer Science ,Methodology (stat.ME) ,010104 statistics & probability ,penalised smoothing ,Linear regression ,Penalties ,overlapping precision matrices ,Applied mathematics ,0101 mathematics ,Variance Parameters ,Smoothing Parameters ,Selection ,Statistics - Methodology ,Components ,Mathematics ,Restricted Maximum-Likelihood ,variance parameter estimation ,Effective Degrees Of Freedomm ,Explained sum of squares ,Random effects model ,Animal-Models ,Computational Theory and Mathematics ,Statistics, Probability and Uncertainty ,Smoothing - Abstract
We present a novel method for the estimation of variance parameters in generalised linear mixed models. The method has its roots in Harville (1977)'s work, but it is able to deal with models that have a precision matrix for the random-effect vector that is linear in the inverse of the variance parameters (i.e., the precision parameters). We call the method SOP (Separation of Overlapping Precision matrices). SOP is based on applying the method of successive approximations to easy-to-compute estimate updates of the variance parameters. These estimate updates have an appealing form: they are the ratio of a (weighted) sum of squares to a quantity related to effective degrees of freedom. We provide the sufficient and necessary conditions for these estimates to be strictly positive. An important application field of SOP is penalised regression estimation of models where multiple quadratic penalties act on the same regression coefficients. We discuss in detail two of those models: penalised splines for locally adaptive smoothness and for hierarchical curve data. Several data examples in these settings are presented., MTM2014-55966-P MTM2014-52184-P
- Published
- 2019