twowaypanel 0.9.5

Econometric Analysis of Nonlinear Panel Models with Individual and Time Effects

twowaypanel: Econometric Analysis of Nonlinear Panel Models with Individual and Time Effects

twowaypanel is a Python package for bias-corrected estimation and inference in nonlinear panel data models with additive individual and time fixed effects.
It accompanies the paper:

Yan, Zizhong, Zhengyu Zhang, Mingli Chen, Jingrong Li, Iván Fernández-Val. (2026), Robust Priors in Nonlinear Panel Models with Individual and Time Effects. arXiv e-prints, arXiv:2604.03663.

Quick links

• Paper: arXiv link
• Documentation (Read the Docs): twowaypanel.readthedocs.io

▶ Overview

Nonlinear panel models with additive individual and time effects are a workhorse in empirical economics. However, fixed-effects maximum likelihood estimation is subject to the incidental parameter problem, which can induce substantial finite-sample bias in both parameter estimates and economically relevant functionals (e.g., average partial effects).

twowaypanel provides bias-corrected estimation and inference for:

• model parameters, and
• average partial effects (APEs), for both continuous regressors and discrete regressors (finite changes).

The current release supports four model classes—binary logit, binary probit, multinomial logit, and ordered logit. The explanatory variables can be either strictly exogenous or predetermined (e.g., include lagged dependent variables to accommodate dynamic models).

▶ Documentation

User-facing documentation (tutorials, examples by model class, and API reference) is maintained on Read the Docs:

• Documentation homepage: twowaypanel.readthedocs.io
• Tutorials: Quick Start and tutorial pages
• Examples gallery: Worked examples by model class

The documentation explains expected data formats, model options, and how to interpret the printed output tables (including APEs and, when relevant, MCMC diagnostics).

▶ Supported models and bias-correction methods

twowaypanel implements the likelihood-based bias-correction methods proposed in Yan et al. (2026), including:

• Integrated-likelihood-based correction (via priors; “prior correction”), and
• Joint-likelihood-based correction (via penalties; “penalty correction”). For logit and probit panels, the package also provides the analytical bias correction for fixed-effects MLE developed by Fernández-Val and Weidner (2016).

Model class	Regressor exogeneity	Generic prior / penalty	Model-specific prior / penalty	Analytical correction (FW16)
Binary logit	Strict exogenous	✓	✓ (Prior SE)	✓
Binary logit	Predetermined	✓	✓ (Prior PE; Prior SE + analytical)	✓
Binary Probit	Strict exog. or predet.	✓	–	✓
Multinomial logit	Strict exogenous	✓	✓ (Prior SML)	–
Multinomial logit	Predetermined	✓	✓ (Prior PML)	–
Ordered logit	Strict exog. or predet.	✓	–	–

Notes.

• Model-specific priors/penalties are defined in Yan et al. (2026) and are designed to deliver improved finite-sample performance within the corresponding model class.
• The generic prior/penalty is intended to be robust across a broad class of nonlinear panel specifications; see Yan et al. (2026) for details and motivation.

▶ Installation

1. Installation from PyPI (recommended)

pip install twowaypanel

2. Install the development version from GitHub

git clone https://github.com/zizhongyan/twowaypanel.git
cd twowaypanel
pip install -e .

3. Install from a local copy

If you downloaded the source code (e.g., as a ZIP file) and unzipped it locally, install from the repository root:

cd /path/to/twowaypanel
pip install -e .

4. Jupyter / notebook users

To ensure the package is installed into the same environment as the current Jupyter kernel, run the following from the repository root directory:

%cd /path/to/twowaypanel
%pip install -e .

▶ Citing this work

Please cite both the underlying paper and the software package when using this code in your research.

1. Paper

   • Yan, Zizhong, Zhengyu Zhang, Mingli Chen, Jingrong Li, Iván Fernández-Val (2026), “Robust priors in nonlinear panel models with individual and time effects.” arXiv e-prints, arXiv:2604.03663.

2. Software

   • Yan, Zizhong, Zhengyu Zhang, Mingli Chen, Jingrong Li, Iván Fernández-Val (2026), “twowaypanel: A Python package for econometric analysis of nonlinear panel models with individual and time effects.“ (Version 0.9.4) [Computer software]. Source code: https://github.com/zizhongyan/twowaypanel.

References

• Fernández-Val, Iván and Martin Weidner (2016), “Individual and time effects in nonlinear panel models with large N, T.” Journal of Econometrics, 192(1), 291–312.

---

Code maintainer: Zizhong Yan, Institute for Economic and Social Research (IESR), Jinan University, Guangzhou, China.
Email: [e-mail]