Multiple imputation of missing data in a dataset using MICT or MICT-timing methods. The core idea of the algorithms is to fill gaps of missing data, which is the typical form of missing data in a longitudinal setting, recursively from their edges. Prediction is based on either a multinomial or random forest regression model. Covariates and time-dependent covariates can be included in the model.
| Version: | 2.2.0 |
| Depends: | R (≥ 3.5.0) |
| Imports: | Amelia,cluster,dfidx,doRNG,doSNOW,dplyr,foreach, graphics,mlr,nnet, parallel,plyr,ranger,rms, stats,stringr,TraMineR,TraMineRextras, utils,mice,parallelly |
| Suggests: | R.rsp,rmarkdown,testthat (≥ 3.0.0) |
| Published: | 2025-01-15 |
| DOI: | 10.32614/CRAN.package.seqimpute |
| Author: | Kevin Emery [aut, cre], Anthony Guinchard [aut], Andre Berchtold [aut], Kamyar Taher [aut] |
| Maintainer: | Kevin Emery <kevin.emery at unige.ch> |
| BugReports: | https://github.com/emerykevin/seqimpute/issues |
| License: | GPL-2 |
| URL: | https://github.com/emerykevin/seqimpute |
| NeedsCompilation: | no |
| Materials: | NEWS |
| CRAN checks: | seqimpute results |