Course 2025-2026 a.y.

Part I – Data Engineering Lab

A. Python Foundations & Environment

• IDEs and notebook workflows (Jupyter, VS Code, remote servers)
• Core Python syntax: variables, data types (numeric, boolean, string), containers (lists, dicts, tuples, sets)
• Control flow: if/elif/else, for and while loops, error handling (try/except)
• Functions and modules: defining, scoping, lambdas, package management

B. Data Acquisition & Import

• Reading and writing CSV, JSON, Excel, and Stata files
• Managing large datasets with chunked loading and memory optimization
• Ensuring data quality at import: type conversions and error reporting

C. Data Wrangling & Cleaning

• Pandas Series and DataFrame fundamentals: indexing, slicing, filtering
• Detecting and treating missing values, duplicates, and outliers
• String operations and date–time parsing
• Grouped aggregations and transformations (groupby, agg, transform)

D. Data Integration & Reshaping

• Merging, joining, concatenating multiple tables (e.g., rounds, valuations, company profiles)
• Pivoting and melting for “wide” vs. “long” formats
• Building event‐time panels and longitudinal structures
• Chaining operations into reproducible pipelines

E. Exploratory Visualization

• Plotting with Matplotlib (line, bar, scatter, box)
• Customizing plots: titles, labels, legends, and annotations
• Exporting figures for reports and presentations

F. Introduction to Text Data

• Regular expressions for parsing term sheets and news feeds
• Tokenization and basic NLP workflows (TF-IDF)

Part II – Econometric Methods

A. Econometric Data Structures

• Cross-section vs. time series vs. panel data
• Constructing balanced and unbalanced panels from transaction logs

B. Simple Linear Regression

• Economic vs. econometric model specification
• Ordinary least squares estimation and interpretation
• Incorporating non-linear terms and interaction dummies

C. Inference & Hypothesis Testing

• Confidence intervals and p-values
• Robust standard errors and heteroskedasticity tests

D. Multiple Regression & Diagnostics

• Multicollinearity, specification tests, omitted-variable bias
• Model selection and goodness-of-fit measures

E. Binary Outcome Models

• Linear probability model, Logit, and Probit for exit or follow-on rounds
• Marginal effects and interpretation

F. Panel Data Techniques

• Pooled OLS, fixed effects, random effects

Hausman test and model choice

• Describe how to structure and harmonize data from multiple sources into tidy, analysis-ready formats.
• Demonstrate proficiency in Python and Stata for data ingestion, cleaning, and management.
• Apply advanced data‐wrangling techniques to detect and correct missing values, outliers, and inconsistencies.
• Transform and merge disparate tables into cross-sectional and panel datasets ready for analysis.
• Interpret the output of regression models within the context of venture financing.
• Evaluate model diagnostics—such as heteroskedasticity tests and endogeneity checks—to ensure valid inference.
• Present empirical results effectively through concise written reports and visualizations.

• Construct end-to-end data workflows to ingest, clean, and prepare complex startup datasets for analysis.

• Design and implement regression models in Python and Stata to test real-world hypotheses.

• Communicate analytical results through clear visualizations and concise written reports.

• Adapt analytical techniques to novel data challenges encountered in professional research, policy, or finance settings.

• Lectures
• Guest speaker's talks (in class or in distance)
• Practical Exercises
• Individual works / Assignments
• Collaborative Works / Assignments

The following methods will be employed:

• Lectures
  • Core concepts in Python, data handling, and econometrics
• Guest Speaker Seminars
  • Talks by industry practitioners on data sources, startup finance, and market insights
• Hands-On Sessions
  • Guided lab sessions focused on coding exercises, data cleaning tasks, and visualization
• Problem-Solving Sessions
  • Instructor-led exercises and self-assessment quizzes to reinforce key techniques
• Group Project Work
  • Team-based empirical research: hypothesis development, data pipeline construction, model estimation, and presentations

Assessment of learning:

1. Written Examination (60%)

• A 90-minute exam at the end of the course covering data engineering and econometric methods.
  • Assesses theoretical understanding and practical problem-solving skills across both parts of the curriculum.

2. Group Project (40%)

• Teams design and execute an empirical study, culminating in an in-class presentation.
  • Evaluation criteria include research design, data pipeline robustness, econometric analysis and presentation effectiveness.
  • Individual contributions and presentation skills are assessed for each team member.

Written Examination (100%)

• A single, 2½-hour individual exam at the end of the course.
• Covers all topics from data preparation through econometric modeling.

Course materials for Part I will consist of tutorials, code notebooks, and reference readings distributed at the start of the term. For Part II, students will receive lecture slides and a selection of research papers in advance. In addition, the econometrics module will draw on specified chapters from:
Hill, R. C., Griffiths, W. E., & Lim, G. C. (2018). Principles of Econometrics. John Wiley & Sons.

Hill, R. C., Griffiths, W. E., & Lim, G. C. (2018). Principles of Econometrics. John Wiley & Sons (selected chapters). Additional readings for part I and II will be specified at the beginning of the course.