IDCeMPy review
A focused open-source Python package for fitting and evaluating inflated choice models.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
IDCeMPy is an open-source Python package for estimating and evaluating discrete-choice models, aimed at analysts working with specialized inflated outcomes. It covers zero-inflated and middle-inflated ordered probit models, including correlated-error variants, as well as generalized-inflated multinomial logit. Standard ordered probit and multinomial logit models are included too. That makes it a fit for Python analysts whose work calls for these model families, rather than a general-purpose statistical analysis environment.
The package’s main strength is its focused modeling workflow. Maximum-likelihood estimation uses numerical optimization, and model evaluation includes AIC, log-likelihood diagnostics, and Vuong model-comparison tests. It also calculates fitted probabilities and marginal effects, so the scope extends from estimation to interpretation and comparison. Inputs and outputs use pandas DataFrames, and the described examples draw on common Python scientific libraries. Analysts already working in that ecosystem can keep data and model work together; those seeking a graphical workflow or a broad collection of unrelated statistical methods should consider a different tool.
IDCeMPy is distributed as open-source software through pip and the project’s GitHub repository, with documentation as its listed support channel. Its stated deployment is desktop, and the product is positioned for solo and small-team use. The package is therefore most appropriate when its specific inflated discrete-choice models match the analysis and a Python-based workflow is acceptable. It is less suited to buyers who need wider platform options, extensive support channels, or statistical software centered on a broader range of methods.
IDCeMPy pros and cons
- Where it wins
- Fits zero- and middle-inflated ordered probit models, including correlated-error variants
- Includes generalized-inflated multinomial logit alongside standard choice models
- Provides likelihood diagnostics, Vuong tests, fitted probabilities, and marginal effects
- Where it doesn't
- Specializes in discrete-choice models rather than broad statistical analysis
- Requires a Python workflow and pandas DataFrame-based inputs and outputs
- Desktop deployment and docs support may not suit teams seeking other environments
IDCeMPy fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update