It’s said that some 70 percent of statistics are made up on the spot (including this one). Data is a critical resource, and in the field of economics, it’s incredibly important not only to be able to understand and present data that support one’s findings, but also to be able to think critically and identify flawed research.
The Introduction to Econometrics course being taught this summer by Senior Lecturer Seyhan Erden at Columbia University provides students with an opportunity to learn from real-world examples to develop core econometrics skills that may be applied to economics research, policy analysis, finance, and business decision-making.
In a recent interview, Erden discusses the wide-ranging uses of econometrics tools, her work in the field, and what students can expect from her Intro to Econometrics class.
Please describe your background and how you got into your field.
I am from Istanbul. My undergrad degree is from Bogazici University and is in economics. Later, I came to the U.S. for graduate studies. I earned my Ph.D. in economics from the University of Wisconsin, where my research centered on tests for non-nested hypotheses. My dissertation explored statistical testing methods, including weighted least squares estimators and the generalized method of moments, using Monte Carlo simulations.
My journey into the field was driven by a deep interest in quantitative methods and their applications in economic analysis. During my graduate studies, I developed a passion for econometric theory and its practical implications, which led me to focus on hypothesis testing and model evaluation. Later, I joined the Department of Economics at Columbia University.
Can you tell us about your specific research interests within economics? Is the nature of your work evolving along with tech developments like AI?
Over time, my work has evolved to include innovative teaching methodologies, integrating technology to enhance learning outcomes in econometrics. In one of the projects, we are developing an application called Metrics Mentor that helps students in econometrics courses better analyze difficult concepts and develop critical thinking and problem-solving skills. This project aims to develop a visual, interactive, graphical online platform to help students understand complex and abstract concepts in econometrics. In another project, we are measuring attitudes of the students toward the core course in econometrics and how attitudes change from the beginning of the semester to the end depending on the methods we use. In another project, we are examining the factors that influence course evaluations. In all my projects, we have some degree of use of AI.
What course are you teaching this summer? Can you tell us about the class and your approach to teaching it?
I am teaching Introduction to Econometrics this summer. It is one of the required core courses in economics and some engineering majors. This course covers multiple regression and related methods for analyzing data in economics and related disciplines. Additional topics include regression with discrete random variables, instrumental variables regression, analysis of random experiments and quasi-experiments, and regression with time series data. Students learn how to conduct—and how to critique—empirical studies in economics and related fields. Accordingly, the emphasis of the course is on empirical applications. We cover each topic using a real-world application. Students do hands-on data analysis using software. I will incorporate Metrics Mentor simulations to help students internalize the key concepts.
What kinds of research interests do your students have? In what settings might they use the economics skills taught in your classes?
My students are interested in various topics related to economics, such as the economics of development, poverty, income distribution, public policy, behavioral economics, global economics, gender issues, environmental economics, big data and machine learning, and financial economics. They can apply the skills they learn in my class to many areas, such as academia and research (pursuing further studies or research positions in economics, data science, or public policy); government and policy analysis (working in agencies such as the Federal Reserve, IMF, or World Bank to analyze economic trends); finance and consulting (using econometric modeling for risk assessment, investment strategies, and business forecasting; tech and data science); applying statistical techniques in roles related to artificial intelligence, machine learning, and business analytics; or education and teaching (implementing innovative teaching methods, like interactive simulations, in economics education).
What major lessons do you hope your students will learn from your Summer Session course?
Students will learn the fundamentals of econometrics and master key concepts like regression analysis, hypothesis testing, and model specification, understanding the assumptions behind econometric models and how violations affect results. Through hands-on experience analyzing economic datasets using statistical software, students will develop the ability to interpret regression output critically and avoid common pitfalls. They will be able to distinguish correlation from causation using tools like instrumental variables, difference-in-differences, and randomized experiments; understand when and why econometric methods can provide reliable policy and business insights; and develop statistical intuition and problem-solving skills––learning how to diagnose and correct issues such as omitted-variable bias, multicollinearity, and heteroskedasticity. On the practical front, they’ll use simulations to see econometric concepts in action; learn how to apply these skills in economics research, policy analysis, finance, and business decision-making; be able to assess the validity of economic claims in academic literature and media; and be able to present their own findings clearly to technical and nontechnical audiences through reports and summaries.
How is teaching a summer course different from teaching in the school year, and what do you enjoy or look forward to most about teaching in the summer?
Summer courses are generally much smaller than fall and spring courses. I usually have about 20 students in the summer as opposed to 100-plus students in the fall and spring. This means each student gets more of my attention, allowing for more personalized interaction, discussions, and hands-on guidance. I know all my students’ names in the summer. Students take one or two classes only in the summer. Without the distraction of multiple simultaneous courses, students can dedicate more energy to mastering econometrics. This can create an intensive but rewarding learning experience. Many summer students enroll because they have a specific academic or professional goal, such as preparing for research or graduate studies, making them highly engaged. My summer course is for 12 weeks, which is not much shorter than the regular semester of 14 weeks. I enjoy teaching in the summer because I get to know each of my students, and each class becomes like a tutoring session. Summer classes are much more personable. Since my summer class is smaller and more interactive, I can provide individualized attention, making it a more engaging and supportive learning experience. The tutoring-style environment likely helps students grasp econometric concepts more effectively and feel more confident applying them.