Data Analytics for Finance
  • Home
  • Lectures
    • Lecture 1
    • Lecture 2
    • Lecture 3
    • Lecture 4
    • Lecture 5
    • Lecture 6
  • Resources
    • Assignment Workflow
    • Stata Cheat Sheet
Author
Affiliation

Caspar David Peter

Rotterdam School of Management, Accounting Department

BM17FI · MSc Finance · Rotterdam School of Management

Data Analytics for Finance

Data analysis in Stata and research methods for empirical finance: from raw data to identification, publication-quality results, and the replication of a published paper.

Schedule and slides Assessment Assignment workflow
Rotterdam School of Management

Dates 3 Feb to 10 Mar 2027

Format 6 sessions on campus, Wednesdays

Grading Assignments 60%, exam 40%

Exam Individual, via ANS, date TBA

Course outline

This course is designed to equip students with comprehensive data analysis skills using Stata and a thorough understanding of research methods in finance. The course emphasizes both theoretical foundations and practical implementation of econometric techniques essential for empirical finance research.

Students will learn how to apply statistical analysis techniques with proper understanding of underlying assumptions, interpret data effectively, and address common methodological challenges in financial research. These skills directly support students’ ability to successfully complete their master thesis replication projects and prepare them for data-intensive roles in the financial industry.

Schedule and slides

The course consists of 6 sessions of 2 hours 45 minutes, held in person on campus. Each session comes with an individual assignment. Assignments 1 to 5 use the Volkswagen Dieselgate scandal as a running example; Assignment 6 is a replication exercise.

Session 1 · Wed 3 Feb 2027

From Messy Data to Research Insights

09:00-11:45 · Theil CB-1

Reproducible project setup, financial databases, tidy data, identifiers and merges. Assignment 1: data wrangling.

Slides

Session 2 · Wed 10 Feb 2027

Data Exploration and Visualization

09:00-11:45 · Van der Goot M2-03 (Sorbonne)

Distributions, relationships, graph design, and publication-quality tables. Assignment 2: data visualization.

Slides

Session 3 · Wed 17 Feb 2027

Regressions and Identification

09:00-11:45 · Theil CB-1

Identification, endogeneity, experiments, and OLS with its assumptions. Assignment 3: regression analysis.

Slides

Session 4 · Wed 24 Feb 2027

De Do Do Do, De Di i Di - Panel Data Methods

11:00-13:45 · Theil CB-1

Good and bad controls, difference-in-differences, fixed effects, clustered standard errors, and staggered adoption. Assignment 4: difference-in-differences.

Slides

Session 5 · Wed 3 Mar 2027

Event Studies and Fama–MacBeth

09:00-11:45 · Van der Goot M2-03 (Sorbonne)

Abnormal and cumulative abnormal returns, benchmark models, and Fama–MacBeth regressions. Assignment 5: event study.

Slides

Session 6 · Wed 10 Mar 2027

Bringing It All Together

13:00-15:45 · Theil CB-1

The full toolkit applied to a published paper, the Dutch replication, and AI in the thesis workflow. Assignment 6: replication exercise.

Slides

Note that the start times of Sessions 4 and 6 differ from the other sessions. Always check the official timetable for last-minute changes.

Learning goals

By the end of this course, students will be able to:

Goal 1

Data Management

Clean, transform, and merge complex datasets using different identifiers (CUSIP, GVKEY, PERMNO) in Stata

Goal 2

Econometric Theory

Understand and verify the assumptions underlying statistical methods used in finance research

Goal 3

Panel Methods

Implement and interpret fixed effects models with appropriate standard error corrections

Goal 4

Causal Inference

Apply difference-in-differences and event studies with proper identification strategies

Goal 5

Research Communication

Create publication-quality tables and essential visualizations for academic papers

Goal 6

Replication Skills

Execute replications of published research with proper documentation and Stata code

Assessment

60%

Individual assignments

  • Six (weekly) assignments throughout the course (10% each)
  • Each assignment must be passed with a grade of at least 50%
  • All assignments must be submitted
  1. Dieselgate: Data Wrangling
  2. Dieselgate: Data Visualization
  3. Dieselgate: Regression Analysis
  4. Dieselgate: Difference-in-Differences
  5. Dieselgate: Event Study
  6. Replication Exercise: Replicating a Published Study with a Twist

40%

Written exam

  • Individual multiple-choice exam on the methods covered in class and the papers discussed or referenced in the lectures
  • Taken via ANS
  • The exam must be passed with a grade of at least 50%
  • Exam date: TBA

Weekly exam preparation quizzes

  • After each lecture, a short Canvas quiz (10 multiple-choice questions) on that lecture and its papers
  • The quizzes are your exam preparation: they show the style and level of the exam questions
  • The quizzes are not graded

Pass requirements: overall course grade ≥ 55%, each assignment grade ≥ 50%, exam grade ≥ 50%, and all assignments submitted.

Materials and tools

Canvas

Canvas is used for announcements, the weekly quizzes, and sharing of resources.

This website

All lecture slides and additional resources. Review the materials before and after each session.

JupyterHub and Stata

Assignments are done in Stata (version 17 or higher) on JupyterHub. You will receive a personal invitation via Canvas. See the assignment workflow and the Stata cheat sheet.

Textbook

Huntington-Klein, N. (2022). The Effect: An Introduction to Research Design and Causality (2nd ed.). Free online at theeffectbook.net; chapters are assigned per session.

Data from financial databases is available through university subscriptions. All software is available through university resources. Basic familiarity with statistics is expected.

BM17FI · Data Analytics for Finance · Rotterdam School of Management, Erasmus University

 

Caspar David Peter · peter@rsm.nl