Package: multipathaic 0.1.3

Michael Obuobi

multipathaic: Multi-Path Stepwise Selection with AIC

Implements a multi-path forward selection algorithm using the Akaike Information Criterion (AIC) for both linear (Gaussian) and logistic (binomial) regression models. Unlike traditional stepwise methods that follow a single path, this approach explores multiple competitive model paths simultaneously by retaining near-optimal models at each step. The package includes stability estimation via bootstrap resampling and identifies plausible models based on AIC tolerance and variable stability thresholds. The information criterion is described by Akaike (1974) <doi:10.1109/TAC.1974.1100705>, and bootstrap resampling by Efron (1979) <doi:10.1214/aos/1176344552>.

Authors:Michael Obuobi [aut, cre]

multipathaic_0.1.3.tar.gz
multipathaic_0.1.3.zip(r-4.7-any)multipathaic_0.1.3.zip(r-4.6-any)
multipathaic_0.1.3.tar.gz(r-4.7-any)multipathaic_0.1.3.tar.gz(r-4.6-any)
multipathaic_0.1.3.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
multipathaic/json (API)

# Install 'multipathaic' in R:
install.packages('multipathaic', repos = c('https://cran.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/mobuobi/multipathaic/issues

On CRAN:

Conda:

3.00 score 6 exports 0 dependencies

Last updated from:658aadf641. Checks:6 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-develOK137
source / vignettesOK287
linux-releaseOK147
windows-develOK112
windows-releaseOK68
wasm-releaseOK94

Exports:build_pathsconfusion_metricslaunch_appmultipath_aicplausible_modelsstability

Dependencies:

Multi-Path AIC Selection: Diabetes Progression Analysis
About the multipathaic Package | Core Functions | Supported Models | Introduction | Load the Package and Data | Feature Engineering | Train/Test Split | Multi-Path AIC Procedure | Algorithm 1: Multi-Path Forward Selection | Algorithm 2: Stability Estimation | Algorithm 3: Plausible Model Selection | Model Evaluation on Test Set | Summary and Parameter Justification | Recommended Parameter Choices | Justification | References
Rendered fromdiabetes-analysis.Rmdusingknitr::rmarkdown

Last update: 2026-10-28
Started: 2026-10-28

Multi-Path AIC Selection: Methodology
Introduction | Problem Setup | Notation | Objective | Algorithm 1: Multi-Path Forward Selection | Motivation | The Multi-Path Approach | Algorithm Details | Input Parameters | Pseudocode | Key Design Choices | Example: Visualization | Algorithm 2: Stability Estimation | Bootstrap Stability | Design Considerations | Example Output | Algorithm 3: Plausible Model Selection | Filtering Criteria | Interpretation of Results | Plausible Models Table | Variable Summary Table | Example | Complete Pipeline: multipath_aic() | Output Structure | Parameter Tuning Guidelines | Quick Reference | Detailed Guidance | Exploration vs. Computation Tradeoff | Problem-Specific Adjustments | Statistical Justification | Why AIC? | Why Delta = 2? | Why Bootstrap Stability? | Why tau = 0.6? | Advantages Over Traditional Methods | vs. Single-Path Forward Selection | vs. All-Subsets Regression | vs. Lasso/Ridge Regression | Practical Recommendations | Workflow | Reporting Results | Limitations and Considerations | Computational Complexity | Model Family | Assumptions | References | Appendix: Mathematical Details | AIC Derivation | Stability Score Estimation | Average Stability for a Model
Rendered frommethod_vignette.Rmdusingknitr::rmarkdown

Last update: 2026-10-28
Started: 2026-10-28