ISEG Summer School 2024

Speaker: Hans Koster, Vrije Universiteit Amsterdam

Date: 8-11 July 2024

Room : NOVO BANCO

Course Descriprion:

Day 1
09:00-10:00	Lecture 1	Spatial Econometrics I (Spatial data)
10:30-11:30	Lecture 2	Spatial Econometrics II (Spatial regressions)
12:00-13:00	Lecture 3	Spatial Econometrics III (Spatial regressions continued)
Day 2
09:00-10:00	Lecture 4	Discrete Choice I (The random utility framework)
10:30-11:30	Lecture 5	Discrete Choice II+III (Estimating discrete choice models)
12:00-13:00	Lecture 6	Identification I (Research design)
Day 3
09:00-10:00	Lecture 7	Identification II (RCTs, OLS, IV)
10:30-11:30	Lecture 8	Identification III (Quasi-experiments, RDD, standard errors)
12:00-13:00	Lecture 9	Hedonic pricing I (Hedonic theory)
Day 4
09:00-10:00	Lecture 10	Hedonic pricing II (Estimating hedonic price models)
10:30-11:30	Lecture 11	Quantitative spatial models I (Model set-up)
12:00-13:00	Lecture 12	Quantitative spatial models II (Estimation)

Speaker: Rita Ginja, University of Bergen

Date: 15-18 July 2024

Room : NOVO BANCO

Course Descriprion:

Health investments

• Grossman M. 2000. “The Human Capital Model.” Chapter 7 in: Culyer and Newhouse (eds.), Handbook of Health Economics, Volume 1 (New York: Elsevier), pp. 347-408. The Human Capital Model – ScienceDirect

• Abrahamsen, S. A., R. Ginja, J. Riise (2021). School Health Programs: Education, Health, and Welfare Dependency of Young Adults. IZA DP No. 14546 School Health Programs: Education, Health, and Welfare Dependency of Young Adults | IZA – Institute of Labor Economics

Socio economics inequalities in Health

• Case, A. and A. Deaton, “Mortality and Morbidity in the 21st Century,” Brookings Papers on Economic Activity, 2018.

• Bütikofer, A., Karadakic, R. and Salvanes, K., 2021, Income Inequality and Mortality: A Norwegian Perspective. Fiscal Studies, 42: 193-221. https://doi.org/10.1111/1475-5890.12261

• Currie, Janet, and Hannes Schwandt. 2016. Mortality Inequality: The Good News from a County-Level Approach. Journal of Economic Perspectives, 30 (2): 29-52.

• Smith J. 1999. “Healthy Bodies and Thick Wallets: The Dual Relationship Between Health and Economic Status.” Journal of Economic Perspectives, 13(2), 145-66. Healthy Bodies and Thick Wallets: The Dual Relation between Health and Economic Status – American Economic Association (aeaweb.org)

• Meghir C, Palme M, Simeonova E. 2018. “Education and Mortality: Evidence from a Social Experiment.”, American Economic Journal: Applied Economics 10(2): 234-256 Practice Styles

• Fadlon, I. and Van Parys, J., 2020. Primary care physician practice styles and patient care: Evidence from physician exits in Medicare, Journal of Health Economics, vol 71.

• Godøy, A., and Dale-Olsen, H., 2018, Spillovers from gatekeeping – Peer effects in absenteeism, Journal of Public Economics, Volume 167, Pages 190-204.

Family health shocks

• Fadlon, I., Nielsen, T.H., 2019. Family health behaviors. American Economic Review 109, 3162–91

Speaker: Niels Johannesen, Oxford University

Date: 22-25 July 2024

Room: NOVO BANCO

Course Descriprion:

In this series of lectures, I will first provide an overview of recent research that advances our ability to measure inequality and understand its causal drivers (Lectures #1-#3). I will then discuss various intersections between inequality and public policy: tax enforcement (Lecture #4), formal and informal insurance (Lecture #5) and monetary policy (Lecture #6).

There will be a small number of slots for participants to present their own research and receive feedback from the other participants and the lecturer. If you are interested in presenting at the summer school, please indicate that in your application and enclose a short description of your project (around 1-2 pages).

1 – Advances in inequality measurement and new descriptive evidence

This lecture provides an overview of recent advances in the measurement of inequality. It discusses long-run trends in income and wealth inequality, distributional national accounts and inter-generational mobility.

• Piketty, T., Saez, E., & Zucman, G. (2018). Distributional national accounts: methods and estimates for the United States. Quarterly Journal of Economics, 133(2), 553-609.

• Saez, E., & Zucman, G. (2016). Wealth inequality in the United States since 1913: Evidence from capitalized income tax data. Quarterly Journal of Economics, 131(2), 519-578.

• Smith, M., Yagan, D., Zidar, O., & Zwick, E. (2019). Capitalists in the twenty-first century. Quarterly Journal of Economics, 134(4), 1675-1745.

• Smith, M., Zidar, O., & Zwick, E. (2023). Top wealth in america: New estimates under heterogeneous returns. Quarterly Journal of Economics, 138(1), 515-573.

• Abramitzky, R., Boustan, L., Jácome, E., & Pérez, S. (2021). Intergenerational mobility of immigrants in the United States over two centuries. American Economic Review, 111(2), 580-608.

• Alesina, A., Hohmann, S., Michalopoulos, S., & Papaioannou, E. (2021). Intergenerational mobility in Africa. Econometrica, 89(1), 1-35.

2 – Advances in the identification of mechanisms and new causal evidence

This lecture discusses new evidence on the mechanisms underlying trends in inequality, including preference heterogeneity, task automation, inheritance, nature and nurture, return heterogeneity.

• Black, S. E., Devereux, P. J., Lundborg, P., & Majlesi, K. (2020). Poor little rich kids? The role of nature versus nurture in wealth and other economic outcomes and behaviours. Review of Economic Studies, 87(4), 1683-1725.

• Nekoei, A., & Seim, D. (2023). How do inheritances shape wealth inequality? Theory and evidence from Sweden. Review of Economic Studies, 90(1), 463-498.

• Epper, T., Fehr, E., Fehr-Duda, H., Kreiner, C. T., Lassen, D. D., Leth-Petersen, S., & Rasmussen, G. N. (2020). Time discounting and wealth inequality. American Economic Review, 110(4), 1177-1205.

• Fagereng, A., Mogstad, M., & Rønning, M. (2021). Why do wealthy parents have wealthy children? Journal of Political Economy, 129(3), 703-756.

• Fagereng, A., Guiso, L., Malacrino, D., & Pistaferri, L. (2020). Heterogeneity and persistence in returns to wealth. Econometrica, 88(1), 115-170.

• Acemoglu, D., & Restrepo, P. (2022). Tasks, automation, and the rise in us wage inequality. Econometrica, 90(5), 1973-2016.

3 – Tax evasion and inequality

Tax returns are the key data source for modern research in inequality by providing comprehensive information about income and sometimes wealth at the level of individuals and households. However, an important limitation of this data source is that tax evasion may distort measurement, potentially in systematic ways. This lecture reviews recent evidence on the distribution of tax evasion, in particular through offshore accounts, and the implications for inequality.

• Alstadsæter, A., Johannesen, N., and Zucman, G. (2019). Tax evasion and inequality. American Economic Review, 109(6), 2073-2103

• Londoño-Vélez, J., & Ávila-Mahecha, J. (2021). Enforcing wealth taxes in the developing world: Quasiexperimental evidence from Colombia. American Economic Review: Insights, 3(2), 131-148.

• Alstadsæter, A., Johannesen, N., Herry, S. L. G., & Zucman, G. (2022). Tax evasion and tax avoidance. Journal of Public Economics, 206, 104587.

• Johannesen, N., Reck, D., Risch, M., Slemrod, J., Guyton, J., & Langetieg, P. (2023). The Offshore World According to FATCA: New Evidence on the Foreign Wealth of US Households. NBER Working Paper 31055.

4 – Taxation, enforcement and inequality at the top

Rising income and wealth shares at the top is major concern for policy. This lecture reviews new evidence on the policies addressing inequality at the top, including wealth taxation and enhanced tax enforcement.

• Hjalte F. Boas, Niels Johannesen, Claus T. Kreiner, Lauge Larsen, and Gabriel Zucman (2023), “Taxing Capital in a Globalized World: The Effects of Automatic Information Exchange”, EU Tax Observatory working paper.

• Jakobsen, K., Jakobsen, K., Kleven, H., & Zucman, G. (2020). Wealth taxation and wealth accumulation: Theory and evidence from Denmark. Quarterly Journal of Economics, 135(1), 329-388.

• Londoño Vélez, J., & Tortarolo, D. (2022). Revealing 21 per cent of GDP in hidden assets: Evidence from Argentina’s tax amnesties (No. 2022/103). WIDER Working Paper.

• Johannesen, N. (2023). The end of bank secrecy: implications for redistribution and optimal taxation. Oxford Review of Economic Policy, 39(3), 565-574.

5- Insurance and inequality at the bottom

Inequality at the bottom is often driven by adverse shocks (e.g. unemployment, disability, divorce) combined with insufficient insurance. This lecture presents evidence on the role of own insurance, social insurance, and informal insurance in cushioning shocks and mitigating the most adverse outcomes.

• Andersen, A. L., Johannesen, N., & Sheridan, A. (2020). Bailing out the kids: new evidence on informal insurance from one billion bank transfers. Working paper.

• Andersen, A. L., Jensen, A. S., Johannesen, N., Kreiner, C. T., Leth-Petersen, S., & Sheridan, A. (2023). How Do Households Respond to Job Loss? Lessons from Multiple High-Frequency Datasets. American Economic Journal: Applied Economics, 15(4), 1-29.

• Stepner, M. (2019). The insurance value of redistributive taxes and transfers. Working paper

• Ganong, P., & Noel, P. (2019). Consumer spending during unemployment: Positive and normative implications. American Economic Review, 109(7), 2383-2424.

• Landais, C., & Spinnewijn, J. (2021). The value of unemployment insurance. Review of Economic Studies, 88(6), 3041-3085.

6 – Distributional effects of monetary policy

While monetary policy is designed to affect economic aggregates such as inflation and GDP growth, it has implications for inequality to the extent that it affects low-income and highincome households differentially. For instance, soft monetary policy may reduce wage inequality by inducing firms to create more jobs for the poor while at the same time increasing wealth inequality by boosting the prices of assets belonging primarily to the rich. This lecture reviews the channels through which monetary policy rates affect households at different positions of the income distribution differentially and provides empirical evidence on the overall distributional effects of monetary policy.

• Andersen, A. L., Johannesen, N., Jørgensen, M., & Peydró, J. L. (2023). Monetary Policy and Inequality. Journal of Finance,

• Amberg, N., Jansson, T., Klein, M., & Picco, A. R. (2022). Five facts about the distributional income effects of monetary policy shocks. American Economic Review: Insights, 4(3), 289-304. Coibion, O., Gorodnichenko, Y., Kueng, L., & Silvia, J. (2017). Innocent Bystanders? Monetary policy and inequality. Journal of Monetary Economics, 88, 70-89

Speaker : Marica Valente, University of Innsbruck

Date : 22 – 25 July 2024

Room : NOVO BANCO

Course Descriprion:

The course provides an in-depth understanding of the foundations, scope and approaches of machine learning for prediction and causal analysis and it focuses on their application to problems in social sciences and economics. Starting from the basics of linear regression, which underlies many machine learning models, this course introduces students to highdimensional predictive and causal problems. In particular, this course provides students with the basic ideas and intuition behind modern machine learning methods as well as an understanding of how, why, and when they work in practice. Students in this course will not only gain a deep understanding of the foundational aspects of machine learning, but they will also acquire the practical skills necessary for their successful applications to real-world problems.

For example, we could download historical weather data to understand what’s the chance that air pollution levels will be dangerously high, given the conditions we expect there to be tomorrow. That’s prediction, but that’s not usually what we care about in many economic or social science applications. In many applications, we care about causal questions. We don’t just care about understanding what would usually happen in a given situation, we want to understand counterfactuals, say, what would happen if we changed the system. For instance, we could ask not only whether tomorrow might have high pollution levels, but whether we can change this by, say, reducing the number of cars on the road. In other words, we are most interested in interventions we might take to improve the situation. As we will see many times in the course, a naive application of machine learning methods that are designed for prediction to causal tasks do not work well. In this course we will see that, if done right, then actually machine learning methods can be usefully adapted to causal tasks, and can be very helpful to the analyst. Thus, the goal of this course is to teach students how ideas from machine learning can be used not only for prediction but also for the study of what-if (thus, causal) questions.

The course will also help students make judgments, and develop an in-depth, critical understanding of the scope and challenges of machine learning and data-driven analytics. Throughout the entire course, students will be invited to assess the strengths and weaknesses of all different methods presented in class. The difference between prediction of observable outcomes versus causal effect estimation of unobservable parameters will be discussed to understand that just correlation is not causation.

Finally, the course will help students improve their communications skills. This course will give the students the possibility to learn how to communicate science, namely, how to present effectively their ideas, findings, proposals, analysis and critical reasoning in the area of data-driven analytics. A special emphasis will be given to oral presentations and pitches in project group works, and to writing scientific papers. 

Contents:

The course will cover the following topics:

Day 1	• Draw differences between Statistics, Econometrics and Machine Learning • Linear Regression, Assumptions, and Flexibility
Day 2	• Machine Learning Methods for Prediction • Non-parametric methods: CART and Random Forests • R session
Day 3	• Machine Learning Methods for Prediction • Parametric methods: LASSO and other regression-based methods • R session
Day 4	• Machine Learning Methods for Causal Analysis • Double Machine Learning and Causal Forests • R session • Applications in Social Sciences and Economics (e.g. Environmental Policy Evaluation, Crime Detection in the Shadow Economy, Drug Costs in the Health Sector)

Learning skills: This course will empower students with the capability to learn several analytical tools and to apply them to real-world problems in an independent and critical way. The content is divided into two Modules: I, II.

Learning objectives:

Module I: Predictive vs. Causal Problems

• Distinguish prediction problems from causal problems

• Describe and justify that correlation is not causation

• Describe why standard statistical methods, such as linear regression, fail in high dimension

• Perform data analysis in R and apply simple predictive machine learning algorithms

• Analyze algorithms’ outputs and compare outputs across methods

Module II: Causal Models and Machine Learning Methods

• Describe how causal models fit into machine learning

• Distinguish how to use machine learning for prediction vs. for causal effect estimation

• Describe the justification behind double machine learning methods

• Describe the justification behind causal forests

• Perform causal effect analysis with machine learning in R

• Analyze applications of causal problems in social science

Methods:

The course consists of lectures complemented by practical sessions.

The exercises in the course will be conducted in R. A brief introduction into R will be provided. However, I recommend that participants familiarize themselves with the software (more details in the Prerequisites section). Please make sure that R and RStudio are installed on your laptops. To download R, go to https://www.r-project.org/, for RStudio, go to https://www.rstudio.com/products/rstudio/download/.

Literature:

Lecture notes and course material will be made available. Inspirational readings (recommended):

• Varian, H. (2014): Big Data: New Tricks for Econometrics. Journal of Economic Perspectives 28(2), pp. 3-28 https://www.aeaweb.org/articles?id=10.1257/jep.28.2.3

• Athey, S. (2018): The Impact of Machine Learning on Economics. The Economics of Artificial Intelligence: An Agenda. University of Chicago Press https://www.gsb.stanford.edu/faculty-research/publications/impact-machine-learning-economics

• Frey, S. & Savage, A. & Torgler, B. (2011). Behavior under Extreme Conditions: The Titanic Disaster. Journal of Economic Perspectives 25(1), pp. 209-22 https://www.aeaweb.org/articles?id=10.1257/jep.25.1.209

The full list of references will be provided in the course. Selected references:

• (textbook) James, G., Witten, D., Hastie, T., and R. Tibshirani, R. (2013): An Introduction to Statistical Learning with Applications in R. Springer.

• (paper) Belloni, A., Chernozhukov, V., and Hansen, C. (2014). High-Dimensional Methods and Inference on Treatment and Structural Effects in Economics. Journal of Economic Perspectives 28(2), pp. 29-50 https://www.aeaweb.org/articles?id=10.1257/jep.28.2.29

• (paper) Wager, S., and Athey, S. (2018): Estimation and Inference of Heterogeneous Treatment Effects using Random Forests. Journal of the American Statistical Association 113(523), pp. 1228-1242 https://www.tandfonline.com/doi/abs/10.1080/01621459.2017.1319839?journalCode

Prerequisites:

No previous knowledge of machine learning is required since this is an introductory class. I expect that students have completed an introduction to statistics or econometrics. The course requires basic knowledge of the linear OLS regression method. Prior experience with R is not a prerequisite, however, it is certainly advantageous.

Therefore, I recommend participants to familiarize themselves with the software using other free online tools, e.g. https://www.datacamp.com/courses/free-introduction-to-r (sign up and start the free course on Introduction to R), https://swirlstats.com/.

Readings:

• [Introductory] Stauffer, R., Chimiak-Opoka, J. Rodríguez-R, L. M., Thorsten, S. Zeileis, A. Introduction to Programming with R. https://discdown.org/rprogramming/

• [Technical] Venables, W. N., Smith, D. M. and the R Core Team (2018): An Introduction to R. https://cran.r-project.org/doc/manuals/r-release/R-intro.pdf

Overview

Content

How does statistics handle low- versus high-dimensional problems? The logical starting point is understanding linear regression, as many machine learning techniques are based on this simple method. The course then shows how linear regression can become flexible and be adjusted to explore high-dimensional associations and causal problems. In addition, I provide an introduction to highly flexible, nonparametric machine learning techniques such as treebased methods. I present their applicability to both predictive and causal problems and draw contrasts with traditional regression approaches.

Estimation of average treatment effects (ATEs) accounting for a large number of variables such as individuals’ socio-economic attributes plays an essential role in modern economics, medicine, and other disciplines by informing policy decision-making or physicians on the effects of their interventions. Several machine learning algorithms have been proposed recently to estimate ATEs in an effective and flexible way by re-purposing predictive machine learning models for causal estimation. In this course I summarize the literature on predictive algorithms and provide concrete guidance for their application for causal effect estimation in high dimensions.

Moreover, targeting policy or treatment interventions to specific subgroups of the population requires the understanding of heterogeneous causal effects. Heterogeneous causal effect analysis focuses on examining individualized treatment effects for individuals or subgroups in a population. Understanding heterogeneous treatment effects can critically guide, for instance, policymakers to identify socioeconomic groups of individuals for which the policy causes the largest effects, design effective policies, and tailor information campaigns for the least responsive groups. In this course I present the recent advancements in the treatment effect and machine learning literature on the estimation of conditional average treatment effects (CATEs) from observational data with binary or continuous treatments.

The guidance I provide is supported by comprehensive R tutorials in which I will carefully explain codes piece by piece and provide tools for autonomous work. This course provides well-documented implementations of different ATE and CATE estimation strategies used in the literature as well as R codes to allow easy use of these methods as well as reproduction of the case studies analysed in class.

Syllabus

MODUL I – Predictive vs. Causal Problems

Statistics, econometrics and machine learning.

• How does econometrics handle low- versus high-dimensional problems? Starting from the basics of linear regression (OLS), this part of the course will introduce students to high-dimensional predictive problems.

• Operational definition(s), motivating empirical facts, the key concepts of ML

Draw contrasts with traditional approaches (OLS in classical statistics)

• The curse of dimensionality for local average estimators and linear regression

• High-dimensional data: Curse or blessing?

How to use machine learning methods for prediction?

Alternative algorithms to linear regression (OLS) that are better suited for prediction are now easily available: This part of the course introduces some of the machine learning algorithms that are most commonly adopted by economists.

Nonparametric methods. Tree-Based Methods

• Classification and Regression Trees, Random Forests (R packages rpart, randomForest, etc.)

Parametric methods. Variable Selection Techniques

• Regression-Based Methods: Lasso.

• Other methods, only scratch surface: Forward Selection, Ridge, Bridge, and Elastic Nets (R packages glmnet, caret, etc.)

R Session: Machine Learning Methods for Prediction

• Prediction of survival in the Titanic Disaster

• For social scientists, evidence about how people behaved as the Titanic sunk offers a quasi-natural field experiment to explore behavior under extreme conditions of life and death

• Using our intuition and individual data on the Titanic Disaster will allow us to predict survival rates and answer some questions: Was it favourable for survival to travel alone or in company? Does one’s role or function (being a crew member or a passenger) affect the probability of survival? Do social norms, such as “Women and children first!” have any effect? Does nationality affect the chance of survival?

Readings:

· Breiman, L. (1996) Heuristics of instability and stabilization in model selection. Ann. Statist., 24, 2350–2383.

· Hoerl, A. and Kennard, R. (1988) Ridge regression. In Encyclopedia of Statistical Sciences, vol. 8, pp. 129–136. New York: Wiley.

· Flom, P. L. and Cassell, D. L. (2007): Stopping stepwise: Why stepwise and similar selection methods are bad, and what you should use. NESUG 2007.

· Varian, H. (2014): Big Data: New Tricks for Econometrics. Journal of Economic Perspectives 28(2), pp. 3-28.

· Giraud, C. (2014): Introduction to High-Dimensional Statistics, Monographs on Statistics & Applied Probability, Chapman & Hall CRC (mathematical foundations of high-dimensional statistics)

· Jones, Z., and Linder, F. (2015): Exploratory Data Analysis using Random Forests.

· Frey, S. & Savage, A. & Torgler, B. (2011). Behavior under Extreme Conditions: The Titanic Disaster. Journal of Economic Perspectives 25(1), pp. 209-22

https://www.aeaweb.org/articles?id=10.1257/jep.25.1.209

· Friedman, J., Hastie, T., and Tibshirani, R. (2008): The Elements of Statistical Learning (Downloadable on Tibshirani website)

· James, G., Witten, D., Hastie, T., and R. Tibshirani, R. (2013): An Introduction to Statistical Learning with Applications in R. Springer.

· Tibshirani, R. (1996) Regression shrinkage and selection via the lasso. J. R. Statist. Soc. B,

58, 267–288

MODUL II – Causal Models and Machine Learning Methods

Linear Regression Methods for Causal Analysis

• Estimation of ATEs using Linear Regression

Machine Learning Methods for Causal Analysis

• Estimation of ATEs: Double Machine Learning

R Session: Machine Learning Methods for Causal Analysis

• Lasso for double machine learning and causal effect estimation (package hdm)

Machine Learning Methods for Heterogeneous Causal Effects

• Intuition behind CATEs and Causal Random Forests

R Session: Machine Learning Methods for Heterogeneous Causal Effects

• Causal forests for CATE estimation (package grf)

Applications

Application fields:

• Applications from Social Sciences and Economics (broadly defined)

• From my research (see https://sites.google.com/view/maricavalente/research): this includes e.g. quantifying crime and irregular migration in the shadow economy, estimating environmental policy effects, and the effects of pharmaceutical payments to physicians in the health sector

Readings:

· Athey, S., Tibshirani, J., and Wager, S. (2018): Generalized Random Forests. Annals of Statistics 47(2), pp. 1148-1178.

· Belloni, A., Chernozhukov, V. (2013). Least squares after model selection in highdimensional sparse models. Bernoulli 19(2), 521-547.

· Belloni, A., Chernozhukov, V., and Hansen, C. (2014). High-Dimensional Methods and Inference on Treatment and Structural Effects in Economics. Journal of Economic Perspectives 28(2), pp. 29-50.

· Belloni, A., Chernozhukov, V., and Hansen, C. (2014). Inference on Treatment Effects After Selection Amongst High-Dimensional Controls,” Review of Economic Studies, 81, 608–650.

· Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C., Newey, W. and Robins, J. (2018), Double/debiased machine learning for treatment and structural parameters. The Econometrics Journal, 21: C1-C68.

· Chernozhukov, V., Hansen, C. and Spindler, M. (2016). High-dimensional Metrics in R. arXiv:1603.01700 · Wager, S., and Athey, S. (2018): Estimation and Inference of Heterogeneous Treatment Effects using Random Forests. Journal of the American Statistical Association 113(523), pp.  1228-1242