| Title: | Automated Routing Engine for Longitudinal Missing Data |
| Version: | 0.1.0 |
| Description: | An automated routing engine for longitudinal missing data. It utilizes a Lagrange-constrained Random Forest based on sample size, missingness rate, and skew to preserve structural variance. |
| License: | MIT + file LICENSE |
| SystemRequirements: | C++17 |
| Encoding: | UTF-8 |
| VignetteBuilder: | knitr |
| Imports: | Rcpp (≥ 1.0.0), missForest, MASS |
| Suggests: | lavaan, ggplot2, tidyr, dplyr, knitr, rmarkdown |
| LinkingTo: | Rcpp, RcppArmadillo |
| Config/roxygen2/version: | 8.0.0 |
| NeedsCompilation: | yes |
| Packaged: | 2026-05-17 22:06:44 UTC; xguo |
| Author: | Xiyuan Guo [aut, cre] |
| Maintainer: | Xiyuan Guo <tommyguo039@gmail.com> |
| Repository: | CRAN |
| Date/Publication: | 2026-05-21 12:30:08 UTC |
Smriti Automated Longitudinal Imputation
Description
This function performs an automated routing and refinement for longitudinal missing data. It establishes a target covariance manifold from observed data, performs initial machine learning imputation, and then projects the result back toward the structural manifold using a Lagrangian constraint.
Usage
smriti_impute(data, time_cols, lambda = 0.5, robust = TRUE)
Arguments
data |
A data frame containing missing values. |
time_cols |
A character vector or numeric vector specifying the longitudinal columns. |
lambda |
A numeric value specifying the penalty weight for the Lagrangian constraint. |
robust |
A logical value. Setting it to TRUE sacrifices a marginal degree of asymptotic efficiency on perfect Gaussian data to secure structural integrity against heavy-tailed skew (the robustness-efficiency tradeoff). |
Value
A data frame with imputed and structurally refined values.