@article{fdi:010076117,
  title = {{B}ridging data exploration and modeling in event-history analysis : the supervised-component {C}ox regression},
  author = {{B}ry, {X}. and {S}imac, {T}. and {E}l {G}hachi, {S}. {E}. and {A}ntoine, {P}hilippe},
  editor = {},
  language = {{ENG}},
  abstract = {{I}n event-history analysis with many possibly collinear regressors, {C}ox's proportional hazard model, like all generalized linear models, can fail to be identified. {D}imension-reduction and regularization are therefore needed. {P}enalty-based methods such as the ridge and the least absolute shrinkage and selection operator ({LASSO}) provide a regularized linear predictor, but fail to highlight the predictive structures. {T}his is the gap filled by the supervised-component {C}ox regression ({SCC}ox{R}). {I}ts principle is to compute a sequence of orthogonal explanatory components, which both rely on the strong correlation structures of regressors and optimize the goodness-of-fit of the model. {O}ne of its parameters tunes the balance between component strength and goodness of fit, thus bridging the gap between classical {C}ox regression with {C}ox regression on principal components. {A} second parameter allows the focus on subsets of highly correlated explanatory variables. {A} third parameter tunes the regularization of the model coefficients, leading to more robust estimates. {S}imulations show how to tune the parameters. {T}he method is applied to the case study of polygamy in {D}akar, {S}enegal.},
  keywords = {{S}upervised components ; {C}ox regression ; partial least squares {C}ox regression ; regularization ; supervised component generalized linear regression ; survival analysis},
  booktitle = {},
  journal = {{M}athematical {P}opulation {S}tudies},
  volume = {27},
  numero = {3},
  pages = {139--174},
  ISSN = {0889-8480},
  year = {2020},
  DOI = {10.1080/08898480.2018.1553413},
  URL = {https://www.documentation.ird.fr/hor/fdi:010076117},
}