Multinomial logistic regression (MNL) with Elastic Net regularization for characterizing LdE-e adoption profiles

This dataset contains the source code and documentation for performing a Multinomial Logistic Regression analysis to identify predictors of adoption profiles for the "Libro del Edificio Electrónico" (LdE-e). The model uses Elastic Net regularization (L1 and L2 penalties) to handle potential multicollinearity and perform variable selection. The analysis includes a global model and stratified models based on climate zones (Cold/Warm) and energy expenditure levels. The scripts are designed for reproducibility in Python environments using standard scientific libraries.