@article{fdi:010061193,
  title = {{N}onlinear multiple imputation for continuous covariate within semiparametric {C}ox model : application to {HIV} data in {S}enegal},
  author = {{M}bougua, {J}. {B}. {T}. and {L}aurent, {C}hristian and {N}doye, {I}. and {D}elaporte, {E}ric and {G}wet, {H}. and {M}olinari, {N}.},
  editor = {},
  language = {{ENG}},
  abstract = {{M}ultiple imputation is commonly used to impute missing covariate in {C}ox semiparametric regression setting. {I}t is to fill each missing data with more plausible values, via a {G}ibbs sampling procedure, specifying an imputation model for each missing variable. {T}his imputation method is implemented in several softwares that offer imputation models steered by the shape of the variable to be imputed, but all these imputation models make an assumption of linearity on covariates effect. {H}owever, this assumption is not often verified in practice as the covariates can have a nonlinear effect. {S}uch a linear assumption can lead to a misleading conclusion because imputation model should be constructed to reflect the true distributional relationship between the missing values and the observed values. {T}o estimate nonlinear effects of continuous time invariant covariates in imputation model, we propose a method based on {B}-splines function. {T}o assess the performance of this method, we conducted a simulation study, where we compared the multiple imputation method using {B}ayesian splines imputation model with multiple imputation using {B}ayesian linear imputation model in survival analysis setting. {W}e evaluated the proposed method on the motivated data set collected in {HIV}-infected patients enrolled in an observational cohort study in {S}enegal, which contains several incomplete variables. {W}e found that our method performs well to estimate hazard ratio compared with the linear imputation methods, when data are missing completely at random, or missing at random.},
  keywords = {multiple imputation ; semiparametric {C}ox model ; {HIV} ; splines function ; linear {B}ayesian regression ; {SENEGAL}},
  booktitle = {},
  journal = {{S}tatistics in {M}edicine},
  volume = {32},
  numero = {26},
  pages = {4651--4665},
  ISSN = {0277-6715},
  year = {2013},
  DOI = {10.1002/sim.5854},
  URL = {https://www.documentation.ird.fr/hor/fdi:010061193},
}