Robizon Khubulashvili

The course is structured to equip students with skills commonly used in the tech sector and empirical work in general. Here's an outline of the course:

1. The course begins with a 3-4 week review covering coding (in Python), matrix operations (linear algebra), statistics, and probability theory.

2. From week 5, I introduce basic economic models, such as investment problems, shortest path problems, etc.

3. Next, we explore simple linear models. The goal is for students to understand estimation as an optimization problem, focusing on identifying parameter values that best fit the data within specified assumptions using Ordinary Least Squares (OLS). I walk through a model in class to derive a closed-form solution and code it live, giving students hands-on coding experience. Toward the end, I provide a package that performs estimations with a single line, ensuring students grasp both the intuition behind the models and their practical implementation.

4. Following this, I introduce maximum likelihood estimation (MLE). After a general introduction and some examples, we revisit the linear model to show how MLE can also be used to find the best-fit line. The focus is on contrasting OLS and MLE, highlighting how they approach the best-fit line from different angles.

5. The final section covers discrete choice models. Here, I aim to help students understand the link between utility functions and choice, key concepts in economics. Using Kenneth Train’s book ("Discrete Choice Methods with Simulation"), we derive a Logit model from agent-based utility maximization, then return to Jupyter Notebooks to manually estimate binary, multinomial, and conditional Logit models.

By the end of the semester, students have engaged extensively in live coding during class, preparing them for coding interviews. They gain an understanding of the role of models in economics, the fundamentals of model estimation, and familiarity with widely used estimation methods such as OLS and MLE.

The course was designed for master's students in economics to gain practical experience with data science topics and work with real-world data. The aim is to provide extensive coding practice while also building a strong understanding of the intuition and mathematical logic behind predictive models. 

1. Understand the fundamentals of predictive modeling and its applications in economics. 

2. Learn how to select and apply appropriate statistical techniques for various predictive modeling problems. 

3. Gain experience using popular predictive modeling tools in Python. 

4. Develop the skills to evaluate and interpret the results of predictive models. 

5. Build the ability to effectively communicate and present findings. 

6. Understand the challenges and limitations of predictive modeling and explore strategies to address them. 

7. Obtain hands-on experience by implementing predictive models on real-world datasets.

This is an advanced microeconomics course designed for master's students in economics who already have some exposure to microeconomic theory. The course primarily focuses on applications in the digital economy, such as auctions, two-sided matching markets, the network economy, and online price discrimination. 

1. We begin with a review of Game Theory, covering how to model strategic interactions and introducing key solution concepts. 

2. After establishing the foundation, students explore auctions in the digital economy, particularly how firms like Google and Facebook auction internet advertising space. 

3. We then introduce two-sided matching markets, using platforms like Airbnb and Uber as examples, as these markets are prevalent in the digital economy. 

4. Finally, we examine network goods and how platforms use user data to engage in price discrimination.

This course introduces microeconomics, focusing on how economic systems determine the production, distribution, and allocation of goods and services. We examine household and firm behavior, market structures, and situations where markets fail to be efficient, exploring potential remedies like taxation and regulation. The goal of the course is for students to start thinking like an economist and understand current economic news.

In my Behavioral Economics course, we explore how psychological factors influence decision-making and challenge the rational agent model from traditional economics. The course aims to improve our understanding of individual choice to better explain economic phenomena.