Journal Article10.1002/CJS.11661
Robust estimation and variable selection for function-on-scalar regression
Xiong Cai,Liugen Xue,Jiguo Cao +2 more
6
TL;DR: Cao et al. as mentioned in this paper developed a robust variable selection procedure for function-on-scalar regression with a large number of scalar predictors based on exponential squared loss combined with the group smoothly clipped absolute deviation regularization method.
read more
Abstract: Function-on-scalar regression is commonly used to model the dynamic behaviour of a set of scalar predictors of interest on the functional response. In this article, we develop a robust variable selection procedure for function-on-scalar regression with a large number of scalar predictors based on exponential squared loss combined with the group smoothly clipped absolute deviation regularization method. The proposed procedure simultaneously selects relevant predictors and provides estimates for the functional coefficients, and achieves robustness and efficiency using tuning parameters selected by a data-driven procedure. Under reasonable conditions, we establish the asymptotic properties of the proposed estimators, including estimation consistency and the oracle property. The finite-sample performance of the proposed method is investigated with simulation studies. The proposed method is also demonstrated with a real diffusion tensor imaging data example. The Canadian Journal of Statistics 00: 000–000; 2021 © 2021 Statistical Society of Canada Résumé: La régression fonction sur scalaire est communément utilisée pour modéliser le lien entre une variable réponse fonctionnelle et un ensemble de prédicteurs scalaires d’intérêt. C’est dans ce cadre que les auteurs de ce travail proposent une procédure de sélection de variables lorsque le nombre de prédicteurs scalaires est grand. La procédure en question est robuste et fait usage de l’exponentielle de l’erreur quadratique couplée avec une méthode de régularisation basée sur une pénalité lisse coupée de la déviation absolue (SCAD). En choisissant des paramètres d’ajustement basés sur les données, cette procédure permet de simultanément sélectionner les prédicteurs pertinents et d’estimer les coefficients fonctionnels de manière robuste et efficace. Le comportement asymptotique des estimateurs proposés, dont la convergence et la propriété d’oracle, est exploré sous des conditions de régularité raisonnables. Quant à leurs propriétés à distance finie, elles sont illustrées à laide de simulations numériques et un exemple pratique de données d’imagerie de diffusion par tenseurs. La revue canadienne de statistique 00: 000–000; 2021 © 2021 Société statistique du Canada Additional Supporting Information may be found in the online version of this article at the publisher’s website. * Corresponding author: jiguo_cao@sfu.ca †Data used in preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List .pdf. © 2021 Statistical Society of Canada 2 CAI, XUE AND CAO Vol. 00, No. 00
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Neural networks for scalar input and functional output
Sidi Wu,Cédric Beaulac,Jiguo Cao +2 more
TL;DR: A feed-forward neural network for scalar input and functional output regression. The model outperforms conventional function-on-scalar regression and scales better with the dimension of the predictors.
Special issue on “Functional and object data analysis”: Guest Editor's introduction
TL;DR: Applied FDA is now a well-established branch of data-oriented nonparametric statistics, and offers numerous visualizations and exploratory tools for functional data, such as functional modes of variation.
3
Bayesian Adaptive Selection of Variables for Function-on-Scalar Regression Models (preprint)
Pedro Henrique T. O. Sousa,Camila P. E. de Souza,Ronaldo Dias +2 more
- 06 Mar 2023
TL;DR: In this article , a new Bayesian method for variable selection in function-on-scalar regression (FOSR) is proposed, which uses latent variables, allowing an adaptive selection since it can determine the number of variables and which ones should be selected.
M-estimation for varying coefficient models with a functional response in a reproducing kernel Hilbert space
Yafei Wang,Bei Jiang,Linglong Kong +2 more
TL;DR: This study proposes an M-estimation framework for varying-coefficient models with functional responses in a reproducing kernel Hilbert space, offering a robust and flexible approach for modeling dynamic relationships in neuroimaging data with smoothness regularization and outlier accommodation.
Bayesian adaptive selection of basis functions for functional data representation
TL;DR: In this paper , the Gibbs sampler is used to select basis functions for a finite representation of functional data, which can deal with observed curves that may differ due to experimental error and random individual differences between subjects, which one can observe in a real dataset application involving daily numbers of COVID-19 cases.
References
Regression Shrinkage and Selection via the Lasso
TL;DR: A new method for estimation in linear models called the lasso, which minimizes the residual sum of squares subject to the sum of the absolute value of the coefficients being less than a constant, is proposed.
Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
Jianqing Fan,Runze Li +1 more
TL;DR: In this article, penalized likelihood approaches are proposed to handle variable selection problems, and it is shown that the newly proposed estimators perform as well as the oracle procedure in variable selection; namely, they work as well if the correct submodel were known.
Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data.
Stephen M. Smith,Mark Jenkinson,Heidi Johansen-Berg,Daniel Rueckert,Thomas E. Nichols,Clare E. Mackay,Kate E. Watkins,Olga Ciccarelli,M Z Cader,Paul M. Matthews,Timothy E.J. Behrens +10 more
TL;DR: TBSS aims to improve the sensitivity, objectivity and interpretability of analysis of multi-subject diffusion imaging studies by solving the question of how to align FA images from multiple subjects in a way that allows for valid conclusions to be drawn from the subsequent voxelwise analysis.
6.6K
Nearly unbiased variable selection under minimax concave penalty
TL;DR: It is proved that at a universal penalty level, the MC+ has high probability of matching the signs of the unknowns, and thus correct selection, without assuming the strong irrepresentable condition required by the LASSO.
3.8K
Nearly unbiased variable selection under minimax concave penalty
TL;DR: In this paper, the authors proposed a penalized linear unbiased selection (PLUS) algorithm, which computes multiple exact local minimizers of a possibly nonconvex penalized loss function in a certain main branch of the graph of critical points of the loss.