Discrete choice modeling in Python with large datasets & single choice / basket models - Assortment & Pricing Optimization .

Computational graph-based discrete choice models

This library simulates the inventory control problem of perishable products by means of Discrete Choice Methods (DCM)

Estimation of Gaussian Process - Latent Class Choice Models (GP-LCCM) using the Expectation Maximization (EM) algorithm

Supplementary Code and Data for "Network Formation and Dynamics among Multi-LLMs"

Estimation of Gaussian Bernoulli Mixture - Latent Class Choice Models (GBM-LCCM) using the Expectation Maximization (EM) algorithm

Python implementation of Multinomial Logit Model

Metropolis-Core: the Rust-based core of the METROPOLIS2 simulator.

This work pertains to detailed analysis, through survey in and around Kolkata region, of the existing Multi-Dimensional Poverty Index used by India to calculate the level of poverty at individual, state and nation level.

Provide a Stated Preference (SP) Survey analysis focusing separately on experimental design, questionnaire creation and SP econometrical analysis

This is an AI-powered modeling framework developed to forecast restaurant demand, classify popularity tiers, and simulate visitor decision-making in urban environments. The project combines LSTM, Kolmogorov–Arnold Networks (KAN), and Discrete Choice Modeling (DCM) to deliver predictive and interpretable insights for smart city planning, operations.