This course provides a detailed coverage of Bayesian inferential methods and their applications to a variety of problems drawn from economics and business. Starting with basic concepts of probability and inference, the treatment covers prior and posterior distributions, classical and MCMC simulation methods, regressions for univariate and multivariate outcomes, and computation of the marginal likelihood and model choice. The key learning objective is for students to develop hands-on Bayesian and Python skills required to conduct data analysis useful for economic and financial decision making. The course will help prepare students entering doctoral education or starting careers in economics, finance, marketing, operations, accounting, political science, statistics, and biostatistics.
- Spring 2024 Syllabus
- Location: Cook Hall 236
- Time: Tuesday 6:00pm-9:15pm
- Office hours: TR 2:00pm-3:00pm & by appointment
- Discord: discord.gg/SsrNPEeP2P
- TA: openai.com/blog/chatgpt
- Zoom: slu.zoom.us/my/econdojo
- Lecture 1: Basic Concepts of Probability and Inference
- Lecture 2: Posterior Distributions and Inference
- Lecture 3: Prior Distributions
- Lecture 4: Classical Simulation
- Lecture 5: Basics of Markov Chains
- Lecture 6: Simulation by MCMC Methods
- Lecture 7: Linear Regression and Extensions
- Lecture 8: Multivariate Responses
- Lecture 9: Time Series
- Python grammatical basics
- Data operations with pandas
- Array operations with numpy
- Classical simulation methods
- MCMC simulation methods
- Application to regression with Student-t error
- Application to Tobit censored regression
- Application to binary response data
- Application to seemingly unrelated regression
- Application to panel data
- Application to regression with autoregressive error
Acknowledgments: development of these lectures has been greatly benefited from discussions with Siddhartha Chib and Hailong Qian. ChatGPT and GitHub Copilot provide excellent teaching assistance on Python programs.