RT Dissertation/Thesis
T1 New covariates selection approaches in high dimensional or functional regression models
A1 Freijeiro González, Laura
K1 High dimension
K1 Covariates selection
K1 Regularization techniques
K1 Distance covariance
K1 Functional concurrent model
AB In a Big Data context, the number of covariates used to explain a variable ofinterest, p, is likely to be high, sometimes even higher than the available sample size (p > n). Ordinary proceduresfor fitting regression models start to perform wrongly in this situation. As a result, other approaches are needed. Afirst covariates selection step is of interest to consider only the relevant terms and to reduce the problemdimensionality. The purpose of this thesis is the study and development of covariates selection techniques forregression models in complex settings. In particular, we focus on recent high dimensional or functional datacontexts of interest. Assuming some model structure, regularization techniques are widely employed alternativesfor both: model estimation and covariates selection simultaneously. Specifically, an extensive and critical review ofpenalization techniques for covariates selection is carried out. This is developed in the context of the highdimensional linear model of the vectorial framework. Conversely, if no model structure wants to be assumed, stateof-the-art dependence measures based on distances are an attractive option for covariates selection. Newspecification tests using these ideas are proposed for the functional concurrent model. Both versions areconsidered separately: the synchronous and the asynchronous case. These approaches are based on noveldependence measures derived from the distance covariance coefficient.
YR 2023
FD 2023
LK http://hdl.handle.net/10347/30892
UL http://hdl.handle.net/10347/30892
LA eng
DS Minerva
RD 24 abr 2026