Preface#
This book is designed to offer clear and comprehensive guidance on causal inference methods, introducing key concepts in this rapidly evolving field. It emphasises the importance of combining a deep understanding of data and processes to draw reliable causal conclusions. Throughout, the book demonstrates how these methodologies can be applied effectively within electricity markets, helping readers to develop practical skills for deriving meaningful insights from data.
Prerequisites#
To fully benefit from this book, it is recommended that readers have a foundational understanding of the following areas:
Statistics and probability: a basic grasp of fundamental statistical concepts and probability theory.
Python programming: a working knowledge of Python, including familiarity with libraries such as NumPy, pandas, statsmodels, and scikit-learn.
Machine learning: an understanding of linear regression models and a general overview of machine learning techniques.
If you find that any of these topics require a refresher, the Appendix provides a dedicated crash course covering these essential concepts. The chapters are structured to allow you to review the foundational material at your own pace. However, if you are already well-versed in these areas or wish to dive directly into the application of causal inference methods, feel free to skip those chapters.
Learning Objectives#
By the end of this book, readers will be able to:
Apply causal inference methodologies to identify and analyse causal relationships between variables.
Utilise directed acyclic graphs (DAGs) to represent causal structures and reason about their implications for statistical modelling.
Implement and interpret causal discovery algorithms, understanding their assumptions, strengths, and limitations.
Estimate causal effects using advanced techniques, such as instrumental variables, double machine learning, and difference-in-differences.
Interpret the outputs of complex machine learning models and identify the key influences of each feature.
Design and conduct experiments to generate data suitable for rigorous causal analysis.
Note
Each chapter can be downloaded as a Jupyter Notebook, facilitating hands-on learning and the easy reproduction of results by allowing you to generate data and apply statistical methods.
The chapters are designed to be as self-contained as possible, enabling you to focus on specific topics without relying on prior chapters (within certain limits).
References
This book is primarily focused on practical applications, offering essential insights and examples for applying causal inference methods to real-world problems. For readers seeking additional theoretical and technical depth, we recommend the following resources: