ubcecon/ECON526

ECON526 - Fall 2023

This is a MA-level course in quantitative economics, data science, and causal inference in economics.

This course will have a combination of coding, theory, and development of mathematical background. All coding is done in Python.

Link to Lecture Slides

and Paul's HTML Slides, source

Course materials

• Get a GitHub ID and apply for the Student Developer Pack to get further free features
• Consider clicking Watch at the top of this repository to see file changes

All materials will be on github, and canvas will be used to submit assignments/communication.

There is no assigned physical textbook, but we will be using lecture notes from:

• Causal Inference for The Brave and True
• QuantEcon Python and Data Science Lectures

Computing Environment

While you can use the UBC JupyterOpen for this course, we strongly suggest installing Python on your local machine. The easiest way to do this is:

• Install Anaconda to install python and its packages for your operating system
• Install git for your operating system
• Optionally: install (a) Github Desktop; (b) VS Code to make it easier to manage downloaded notebooks.
• Then clone the following repositories onto your local machine using a terminal, using either git directly (e.g. in terminal go git clone https://github.com/ubcecon/ECON526.git), GitHub Desktop, or VS Code
  • https://github.com/ubcecon/ECON526
  • https://github.com/QuantEcon/lecture-python-intro.notebooks
  • https://github.com/QuantEcon/lecture-python.notebooks
  • https://github.com/QuantEcon/lecture-datascience.notebooks
  • In some cases you will need to manually install packages (by doing pip install -r requirements.txt within some of those repositories, or manually installing packages as required)

For introductory users, we recommend using GitHub Desktop which allows cloning from a button on the public github repo directly. For intermediate users, we recommend skipping the GitHub Desktop and instead using VS Code since you will likely begin using the VSCode editor as your primary Python (and latex) editor sooner than later.

Syllabus

See Syllabus for more details

Problem Sets and Exams

The course has two midterms, weekly to bi-weekly problem sets, and a final data project due the last day of class.

1. September 10th Midnight: Problem Set 1
2. September 17th Midnight: Problem Set 2
3. September 25th Midnight: Problem Set 3
4. October 4th Midnight: Problem Set 4
5. October 5th: Midterm Logistics Practice with Midterm Practice Problems
6. October 11th: IN CLASS MIDTERM #1'
7. October 25th: Problem Set 5
8. October 2nd, 10am: Problem Set 6
9. November 8th: IN CLASS MIDTERM #2
10. December 8th Midnight: Data Project Due

See the /problem_sets folder within this repository for the problem sets as jupyter notebooks. You should modify them directly as Jupyter notebooks, and the TA will explain how to submit them.

Lectures

This year the course will be taught in three parts where the later parts of the course will follow material in Causal Inference for The Brave and True.

This lecture begins assuming you have completed the math/programming bootcamp for our masters students, or had an existing python-based programming course. To refresh your knowledge, see basics in QuantEcon Data Science Lectures or QuantEcon Python Programming for Economics and Finance.

Slides for the lectures can be found here and

Paul's HTML Slides, source

Jesse

• September 6: Introduction to Numerical Linear Algebra and its Applications in Data Science

  • Topics: Overview of computational complexity and numerical precision, solving systems of equations, geometric interpretations of linear algebra, matrix decompositions, linear least squares, and eigenvalues and eigenvectors. Preparation for applications.
  • Material:
    • Linear Algebra Foundations, Jupyter, PDF
  • Self-study:
    • Basics of linear algebra, matrices, norms, and linear independence: https://python.quantecon.org/linear_algebra.html
    • Numerical optimization: https://datascience.quantecon.org/scientific/optimization.html
    • Systems of Equations: https://python.quantecon.org/linear_algebra.html#solving-systems-of-equations
    • Eigenvectors and eigenvalues: https://python.quantecon.org/linear_algebra.html#eigenvalues-and-eigenvectors
    • Downloading and manipulating data in Python: https://intro.quantecon.org/long_run_growth.html and https://intro.quantecon.org/business_cycle.html
  • (Optional) Extra Material:
    • Introductory material on linear algebra: https://intro.quantecon.org/linear_equations.html and https://datascience.quantecon.org/scientific/applied_linalg.html
    • Matrix decompositions and other topics: https://python.quantecon.org/linear_algebra.html#further-topics

• September 11: Applications of Linear Algebra (Eigenvalues and Discounting)

  • Topics: Geometric series and present values, difference equations, steady states, and convergence, unemployment dynamics, present discounted values
  • Material:
    • Finishing off Linear Algebra Foundations
    • Eigenvalues and Stability, Jupyter, PDF
  • Self-study:
    • Geometric Series and Present Values: https://intro.quantecon.org/geom_series.html#example-interest-rates-and-present-values
    • Portfolio example: https://datascience.quantecon.org/scientific/applied_linalg.html#portfolios
    • Unemployment Dynamics example: https://datascience.quantecon.org/scientific/applied_linalg.html#unemployment-dynamics
  • (Optional) Extra Material:
    • Supply and Demand: https://intro.quantecon.org/intro_supply_demand.html
    • More on Competitive Equilibrium: https://intro.quantecon.org/supply_demand_multiple_goods.html

• September 13: Latent Variables and Intro to Unsupervised Learning

  • Topics: Finished off eigenvalues and dynamics, principle components, and present discounted values
  • Material:
    • Latent Variables and Unsupervised Learning, Jupyter, PDF
  • Self-study:
    • scikit-learn PCA docs
    • seaborn tutorials

• September 18: More on Latent Variables and Clustering

  • Topics: Finish off continuous latent variables, PCA, auto-encoders, clustering, and started dynamics
  • Material:
    • Latent Variables and Unsupervised Learning
  • Self-study:
    • scikit-learn k-means docs

• September 20: Dynamics

  • Topics: Dynamical systems, stability, fixed points, linearization, intro to the Solow-Swan growth model
  • Material:
    • Linear and Nonlinear Dynamics, Jupyter, PDF
  • Self-study:
    • Solow-Swan Growth Model Derivation: https://intro.quantecon.org/solow.html (skip 20.3)
    • Nonlinear Dynamics and Stability: https://intro.quantecon.org/scalar_dynam.html
    • Review taylor series, just to first order
  • (Optional) Extra Material:
    • More on the Solow Model and Python: https://python-programming.quantecon.org/python_oop.html#example-the-solow-growth-model

• September 25: Probability, Randomness, and Independence

  • Topics: Axioms of probability, LLN and CLT, and Conditional Independence
  • Material:
    • Probability, Conditioning, and Independence, Jupyter, PDF
  • Self-study:
    • https://datascience.quantecon.org/scientific/randomness.html
    • https://intro.quantecon.org/prob_dist.html
  • (Optional) Extra Material:
    • https://python.quantecon.org/lln_clt.html
    • https://python.quantecon.org/prob_meaning.html
    • https://python.quantecon.org/prob_matrix.html

• September 27: Stochastic Processes and Forecasts

  • Topics: Conditional expectations, Bayes' rule, Law of Iterated Expectations, stochastic processes
  • Material:
    • Finish Probability, Conditioning, and Independence
    • Stochastic Processes, Jupyter, PDF
  • (Optional) Extra Material:
    • https://python.quantecon.org/finite_markov.html for more on Markov Chains
    • https://python.quantecon.org/ar1_processes.html for more on AR(1) processes
    • https://datascience.quantecon.org/scientific/randomness.html#loan-states for a simple Markov Chain example

• October 2 (Statutory holiday)

• October 4: Markov Chains and Introduction to Causality and Counterfactuals

  • Topics: Finish stochastic processes and Markov Chains and briefly setup causality and counterfactuals
  • Material:
    • Finish Stochastic Processes
  • Self-Study:
    • https://matheusfacure.github.io/python-causality-handbook/01-Introduction-To-Causality.html

• October 9 (Statutory holiday)

• October 11

  • IN CLASS MIDTERM #1

Phil

• October 12 ("Make-up Monday") Introduction to Causality, continued

  • Material:
    • Introduction to Causality and Randomized Experiments
  • Self-study:
    • https://matheusfacure.github.io/python-causality-handbook/01-Introduction-To-Causality.html
    • https://matheusfacure.github.io/python-causality-handbook/02-Randomised-Experiments.html

• October 16 Stats Review: Quantifying Uncertainty in Causal Inference

  • Material:
    • Stats Review: Quantifying Uncertainty in Applied Economics
  • Self-study:
    • https://matheusfacure.github.io/python-causality-handbook/03-Stats-Review-The-Most-Dangerous-Equation.html

• October 18 Uncertainty Quantification II

  • Material:
    • Finish Stats Review: Quantifying Uncertainty in Applied Economics

• October 23 Causal Graphical Models

  • Material:
    • Causal Graphical Models
    • https://matheusfacure.github.io/python-causality-handbook/04-Graphical-Causal-Models.html
  • Self-study:
    • https://theeffectbook.net/ch-DrawingCausalDiagrams.html

• October 25 Regression, Confounders

  • Material:
    • https://matheusfacure.github.io/python-causality-handbook/05-The-Unreasonable-Effectiveness-of-Linear-Regression.html
    • https://matheusfacure.github.io/python-causality-handbook/06-Grouped-and-Dummy-Regression.html
    • https://matheusfacure.github.io/python-causality-handbook/07-Beyond-Confounders.html

• October 30 Instrumental Variables, Noncompliance, LATE

  • Material:
    • https://matheusfacure.github.io/python-causality-handbook/08-Instrumental-Variables.html
    • https://matheusfacure.github.io/python-causality-handbook/09-Non-Compliance-and-LATE.html

• November 1: Matching

  • Material:
    • https://matheusfacure.github.io/python-causality-handbook/10-Matching.html
    • https://matheusfacure.github.io/python-causality-handbook/11-Propensity-Score.html

Paul

HTML Slides, source

• November 6: Matching
• November 8
  • IN CLASS MIDTERM #2
• November 9 (TA session): Predictive Models
  • https://matheusfacure.github.io/python-causality-handbook/17-Predictive-Models-101.html
  • https://datascience.quantecon.org/tools/regression.html
• November 13 (Midterm Break)
• November 15 (Midterm Break)
• November 20: Finish Matching, start Difference in Differences
  • https://matheusfacure.github.io/python-causality-handbook/13-Difference-in-Differences.html
• November 22: Panel Data and Fixed Effects, The Difference in Differences Saga
  • https://matheusfacure.github.io/python-causality-handbook/14-Panel-Data-and-Fixed-Effects.html
  • https://matheusfacure.github.io/python-causality-handbook/24-The-Diff-in-Diff-Saga.html
• November 27: Synthetic Control
  • https://matheusfacure.github.io/python-causality-handbook/15-Synthetic-Control.html
  • https://matheusfacure.github.io/python-causality-handbook/25-Synthetic-Diff-in-Diff.html
• November 29: Debiased/Orthogonal Machine Learning
  • https://datascience.quantecon.org/applications/ml_in_economics.html#estimation-of-nuisance-functions
• December 4: Treatment Heterogeneity and Conditional Effects
  • https://datascience.quantecon.org/applications/heterogeneity.html
• December 6: 8. Neural Networks
  • Reading: QuantEcon Datascience: Regression - Neural Networks
• December 15
  • DATA PROJECT DUE