Skip to content

Tutorials & Notebooks

PanelBox provides 193 interactive Jupyter notebooks organized across 15 tutorial categories, covering every aspect of panel data econometrics. Each notebook runs directly in Google Colab with zero setup required.

Quick Start

New to PanelBox? Start with Fundamentals to learn the basics, then explore the specific technique you need.

Learning Paths

Choose a path based on your goals and experience level:

Path Level Duration Topics Notebooks
Essentials Beginner 4--6 hours Fundamentals, Static Models, Standard Errors ~18
Applied Researcher Intermediate 12--16 hours + GMM, Discrete Choice, Diagnostics, Visualization ~50
Econometrician Advanced 30--40 hours + Spatial, Quantile, VAR, Frontier, all models 193

Essentials Path (4--6 hours)

For researchers new to panel data or PanelBox. Covers the foundations you need for any applied work.

  1. Fundamentals -- Panel data concepts, within/between variation (4 notebooks)
  2. Static Models -- Pooled OLS, Fixed Effects, Random Effects (7 notebooks)
  3. Standard Errors -- Robust inference basics (notebooks 01--02)

Applied Researcher Path (12--16 hours)

For researchers ready to apply PanelBox to real-world problems with proper diagnostics.

  1. Complete the Essentials path first
  2. GMM -- Dynamic panel models (notebooks 01--04)
  3. Discrete Choice -- Binary, ordered, multinomial (notebooks 01--04)
  4. Validation & Diagnostics -- Testing assumptions (4 notebooks)
  5. Visualization & Reports -- Charts and HTML reports (4 notebooks)

Econometrician Path (30--40 hours)

The complete PanelBox curriculum. Master every model family and technique.

  1. Complete the Applied Researcher path first
  2. Spatial Econometrics -- SAR, SEM, SDM (8 notebooks)
  3. Quantile Regression -- Beyond-the-mean analysis (10 notebooks)
  4. Panel VAR -- Vector autoregressions (7 notebooks)
  5. Stochastic Frontier -- Efficiency analysis (6 notebooks)
  6. Count Models -- Poisson, PPML, zero-inflated (7 notebooks)
  7. Censored & Selection -- Tobit, Heckman (8 notebooks)
  8. Marginal Effects -- Interpretation of nonlinear models (6 notebooks)

Tutorial Categories

  • Fundamentals

    Panel data basics, formulas, estimation & interpretation

    4 notebooks | Beginner -- Intermediate

    Fundamentals

  • Static Models

    Pooled OLS, Fixed Effects, Random Effects, IV estimation

    7 notebooks | Beginner -- Advanced

    Static Models

  • Dynamic GMM

    Arellano-Bond, Blundell-Bond, CUE, bias correction

    6 notebooks | Intermediate -- Advanced

    GMM

  • Spatial Econometrics

    SAR, SEM, SDM, dynamic spatial panels, marginal effects

    8 notebooks | Intermediate -- Advanced

    Spatial

  • Stochastic Frontier

    SFA, four-component model, TFP decomposition

    6 notebooks | Intermediate -- Advanced

    Frontier

  • Quantile Regression

    Panel quantile methods, Canay, location-scale, QTE

    10 notebooks | Intermediate -- Advanced

    Quantile

  • Panel VAR

    VAR, VECM, IRF, FEVD, Granger causality

    7 notebooks | Intermediate -- Advanced

    VAR

  • Discrete Choice

    Logit, probit, ordered, multinomial, dynamic models

    9 notebooks | Beginner -- Advanced

    Discrete

  • Count Data

    Poisson, negative binomial, PPML, zero-inflated models

    7 notebooks | Beginner -- Advanced

    Count

  • Censored & Selection

    Tobit, Honore estimator, Heckman selection models

    8 notebooks | Beginner -- Advanced

    Censored

  • Standard Errors

    Robust, clustered, HAC, Driscoll-Kraay, bootstrap

    7 notebooks | Beginner -- Advanced

    Standard Errors

  • Marginal Effects

    AME, MEM for discrete, count, and censored models

    6 notebooks | Beginner -- Advanced

    Marginal Effects

  • Validation & Diagnostics

    Assumption tests, unit roots, cointegration, specification

    8 notebooks | Intermediate -- Advanced

    Validation

  • Visualization

    Interactive charts, visual diagnostics, automated reports

    4 notebooks | Beginner -- Intermediate

    Visualization

  • Production

    Prediction, model persistence, pipelines, versioning

    6 notebooks | Intermediate -- Advanced

    Production

How to Use These Tutorials

Click the Open in Colab badge on any tutorial to launch it directly in Google Colab. PanelBox will be installed automatically in the first cell.

!pip install panelbox

No local setup required -- just a Google account and a web browser.

Clone the repository and run notebooks locally:

git clone https://github.com/PanelBox-Econometrics-Model/panelbox.git
cd panelbox
pip install -e .
jupyter lab examples/

Browse notebooks on GitHub and download what you need:

https://github.com/PanelBox-Econometrics-Model/panelbox/tree/main/examples/

Difficulty Levels

Level Description Prerequisites
Beginner No prior panel data experience required. Covers fundamentals and basic models. Python, pandas, basic statistics
Intermediate Assumes basic panel data knowledge. Introduces advanced techniques and diagnostics. Fundamentals tutorials completed
Advanced For experienced users. Complex models, custom workflows, and production use cases. Multiple tutorial categories completed

Interactive Notebook Tutorials

In addition to the example notebooks, PanelBox includes self-contained tutorial notebooks covering specific topics:

Tutorial Topic Level
Panel Quantile Regression Introduction to panel quantile methods Intermediate
Panel Cointegration Cointegration testing for panel data Advanced
Multinomial Logit Multinomial choice modeling Intermediate
Panel Unit Root Unit root testing for panels Intermediate
Stochastic Frontier SFA fundamentals and estimation Intermediate
J-Test Specification J-test for non-nested models Advanced
PPML Gravity Gravity models with PPML Intermediate
Spatial Econometrics Complete spatial analysis Advanced

Solutions & Answer Keys

Most tutorial categories include complete solution notebooks. These provide:

  • Full code implementations for all exercises
  • Detailed interpretation of results
  • Best practices and common pitfalls

Look for the Solutions section on each category page.

What's Next?

Start with Fundamentals if you are new to panel data, or jump directly to the category that matches your research needs. Each tutorial page includes recommended learning paths and prerequisites.

See Also