tooluniverse-statistical-modeling

What problem does it solve?

Enable reproducible, publication-ready statistical modeling for biomedical and clinical datasets by providing clear workflows to select, fit, and interpret linear, logistic, ordinal, mixed-effects, and survival models. Reduce errors and save time by standardizing model selection, diagnostics, and formatted reporting so analysts can focus on interpretation and decision-making.

Core Features & Use Cases

• Model selection & fitting: linear regression, binary/ordinal/multinomial logistic, mixed-effects models, Cox proportional hazards, Kaplan-Meier.
• Effect estimation: compute odds ratios, hazard ratios, confidence intervals, p-values, and percentage change comparisons between crude and adjusted models.
• Model diagnostics & formatting: automated diagnostic checks (VIF, Breusch-Pagan, Shapiro-Wilk, proportional hazards checks), per-feature ANOVA pipelines for genomic data, and publication-ready summary tables.
• Real-world use case: fit an ordinal logistic model for COVID-19 severity, extract the OR for vaccination with 95% CI and p-value, run proportional-odds checks, and produce a formatted table for manuscript inclusion.

Quick Start

Fit the appropriate regression (e.g., logistic, ordinal, Cox, mixed-effects) on my dataset and return effect sizes (OR/HR), 95% confidence intervals, p-values, diagnostics, and a publication-ready summary.