Python Workshop

Python Coding for Accounting and Finance Research

This workshop is part of the AFAANZ Doctoral Education Network (AFDEN) program, designed for doctoral students and early-career researchers in accounting and finance who want to leverage Python for empirical research.

Overview

This course is designed to equip doctoral students and early career researchers in accounting and finance with foundational and practical skills in Python programming, tailored specifically for empirical research. I will cover core programming concepts, data manipulation techniques, data collection from databases and the web, natural language processing techniques, and modern statistical and machine-learning methods using Python. Through hands-on assignments and coding exercises, students will learn how to efficiently collect and clean data, perform statistical testing, build predictive models, and visualize results. The goal is not only to become competent in coding but also to use Python as a tool to enhance the quality and efficiency of academic research.

A distinctive feature of this workshop is that almost every module and session is built around understanding or replicating a published study from a top journal, such as Sloan’s (1996, The Accounting Review) accrual anomaly, Kim et al.’s (2016 Journal of Accounting and Economics) option volatility smirk, Obaid and Pukthuanthong’s (2022 Journal of Financial Economics) image sentiment, Chen et al.’s (2022, Journal of Accounting Research) earnings forecasting, de Kok’s (2025 Management Science) detection of “non-answers” in earnings call, among many others. Emphasis will be placed on applications to real-world accounting and finance data, such as databases in WRDS, SEC filings, and textual data (e.g., 10-Ks and earnings-call transcripts). Thus, the Python skills you learn are grounded in the same empirical research you’ll be expected to produce.

The course consists of two 2-hour online modules, which introduce the Python fundamentals for research (Module 1) and the pandas package for handling and analysing data (Module 2), followed by a two-day in-person workshop: Day 1 covers data collection from WRDS and web scraping (Module 3), together with natural language processing and textual analysis (Module 4), and Day 2 covers generative LLMs as a research tool (Module 5) and modern statistical and machine-learning methods for accounting and finance research, including regression, classification, prediction and explainable AI (Module 6). See the GitHub repository for details.

This workshop is offered in September every year. To register, see the information in AFDEN website.

Topics Covered

• Python fundamentals for data analysis
• Data cleaning and manipulation with pandas
• Data visualization with matplotlib and seaborn
• Data collection from WRDS platform using wrds package
• Web scraping for financial data using request and BeautifulSoup
• Incorporation of AI into research workflows
• Text analysis and natural language processing for accounting research using re and nltk
• Regression analysis
• Machine learning basics

Target Audience

Doctoral students and and early-career researchers in accounting and finance. No prior programming experience is required but prior experience in archival accounting or finance research is preferred.

For workshop materials or inquiries, please contact leye.li@unsw.edu.au.