Logistic Curve as a Resolution Framework: A Clean Comparison Against the Classical Exponential Model

# Fractal Logistic Model — Comparison with Classical Exponential We propose the logistic growth model as an operational tool when processesexhibit self-limiting expansion. We present its mathematical formulation, providea head-to-head comparison against the classical exponential model using informa-tion criteria (AIC/BIC) and rolling-origin cross-validation, and include a workedexample with a fitted figure. This repository contains LaTeX, code, data, and figures to evaluate the logistic curve as a resolution framework against the classical exponential model. ## Structure- `doc/` — LaTeX source (`logistic_vs_exp.tex`)- `figs/` — Figures (logistic fit)- `data/` — Input dataset (`dataset.csv`)- `results/` — Output metrics (`metrics_comparison.csv`)- `src/` — Python script (`fit_logistic.py`) ## Quick start```bashpip install numpy pandas matplotlib scipypython src/fit_logistic.py```This will update `results/metrics_comparison.csv` and regenerate the figure in `figs/`. ## Current head-to-head metrics- Exponential: RSS = 11788085.83, BIC = 348.41, CV-MAE = 859.65- Logistic: RSS = 4510072.69, BIC = 326.69, CV-MAE = 586.23 ## LicenseMIT License.