Master's Degree in Economics and Finance

Master's Degree in Specialized Economic Analysis

Master's Degree in Data Science

Admissions

Career Services

Alumni Community

Economics Program

Financial Economics Program

PhD Economics Track Program

Go to Master's degrees

Close

Macroeconometrics

Forecasting with Machine Learning

An introduction to machine learning methods specifically designed for economic time-series forecasting.

17.5h (5 days)

€1,470

Face-to-face

English

Next edition: July 12-16, 2027

Early Bird deadline: April 15, 2027

Learn more

Macroeconometrics

Forecasting with Machine Learning

Complete your details and we'll keep you informed about the next edition

First Name*

Last Name*

Email Address*

Message

I give my consent to be contacted by BSE with information about related academic offers in the future.

Who is this Course for?

Forecasting with Machine Learning

Course overview

This course introduces machine learning methods for forecasting macroeconomic and financial time series. Over five days it moves from regularized regression, through nonlinear and nonparametric methods and large-scale multivariate time-series models, to deep learning and the use of text as data. Each method is developed from first principles and connected explicitly to the classical and Bayesian econometric tools participants already know, such as shrinkage viewed as a prior and neural networks viewed as regressions with hidden stages, before being extended to its modern machine learning counterpart.

Throughout, the emphasis is on what works for economic data: validating models with time-series cross-validation, choosing between sparse and dense representations, understanding why flexible methods deliver smaller gains in macroeconomic forecasting than in the cross-section of asset returns, and interpreting machine learning forecasts through variable importance, SHAP values, and partial dependence. Applications include inflation and output forecasting with large macroeconomic panels, financial conditions indices, empirical asset pricing, and the analysis of central bank communication.

Participants are not required to have prior experience in machine learning.

Faculty

Who is this course for?

This course is useful for:

PhD students and academic researchers incorporating modern prediction methods into their research toolkit, particularly those working on forecasting, asset pricing, or macroeconomic applications
Central bank economists and policy researchers responsible for implementing or evaluating nowcasting and forecasting systems for policy decision-making
Quantitative analysts and finance professionals seeking to enhance forecasting capabilities beyond traditional factor models, particularly those in asset management, risk management, or trading roles
Private sector economists at consulting firms, corporations, and financial institutions requiring accurate forecasting for strategic planning and business decisions
Data scientists transitioning into economic and financial applications who need to understand the unique challenges and best practices for economic time series data

Learning outcomes

By the end of the course, participants will be able to:

Explain the bias–variance trade-off and apply ridge, LASSO, and elastic net regression to forecasting with many predictors, selecting tuning parameters by time-series cross-validation
Apply and interpret nonlinear methods, including kernel methods, Gaussian processes, random forests and gradient boosting, using variable importance, SHAP values and partial dependence
Estimate and forecast with large multivariate models, including Bayesian VARs, factor models and FAVARs, time-varying parameter VARs and Bayesian compressed VARs
Design, train and regularize neural networks for economic and financial prediction, and assess when deep learning adds value over simpler benchmarks
Construct text-based indicators from economic and financial documents, from dictionary-based sentiment and topic models to transformer-based models, and incorporate them into forecasting models
Evaluate and combine competing forecasts out of sample, and select appropriate methods for a given forecasting problem

Key topics for Forecasting with Machine Learning course

Here is a brief outline of what we will cover.

Day 1: Regularized Regression and High-Dimensional Forecasting

Why OLS fails with many predictors: the bias–variance trade-off
Ridge, LASSO and elastic net: geometry, Bayesian interpretation and coefficient paths
Tuning-parameter selection and cross-validation for time series
The sparsity debate: dense versus sparse models; post-selection inference and double selection
Time-varying parameters via shrinkage

Day 2: Nonlinear and Nonparametric Methods

Feature expansions, kernels, kernel regression, and Gaussian processes
Decision trees and pruning
Bagging, random forests, gradient boosting and XGBoost
Hybrid machine learning and econometric models: the macroeconomic random forest and time-varying parameters
Interpretation: variable importance, SHAP values, and partial dependence; evidence on nonlinearity in macroeconomic data

Day 3: Multivariate Time Series and Large-Scale Models

VAR foundations: estimation, forecasting, impulse responses and overparameterization
Bayesian VARs, the Minnesota prior and large BVARs; hierarchical shrinkage priors
Factor models, diffusion indices, targeted predictors and FAVARs; state-space models and the Kalman filter
Large time-varying parameter VARs, stochastic volatility and dynamic model averaging; financial conditions indices
Bayesian compressed VARs and nonlinearity in multivariate models

Day 4: Deep Learning for Economics and Finance

Feedforward neural networks as regressions with hidden stages; universal approximation
Training: backpropagation, stochastic gradient descent and adaptive optimizers
Regularization of neural networks: weight decay, dropout, early stopping and ensembling; overparameterization and when complexity pays off
Neural networks for macroeconomic forecasting; recurrent networks and LSTMs for time series
Deep learning for asset pricing: return prediction, autoencoder factor models and deep stochastic discount factor models

Day 5: Text as Data, Frontiers and Course Synthesis

Text preprocessing and representation: tokenization, bag-of-words and TF-IDF
Dictionary-based sentiment analysis and word embeddings
Topic models and central bank communication
Transformers, attention, BERT and FinBERT, and large language models for economics
Text-augmented forecasting and nowcasting
Frontiers and synthesis: forecast combination, conformal prediction, Bayesian neural networks and method selection in practice

List of References

Core Texts

Core references include foundational texts in statistical learning and deep learning, alongside recent research published in leading econometrics, statistics and finance journals, and survey articles covering recent advances in machine learning for economic forecasting.

Key readings include:

Hastie, T., Tibshirani, R. and Friedman, J. (2009). The Elements of Statistical Learning. Springer.
Goodfellow, I., Bengio, Y. and Courville, A. (2016). Deep Learning. MIT Press.
Korobilis, D. and Shimizu, K. (2022). Bayesian approaches to shrinkage and sparse estimation. Foundations and Trends in Econometrics.
Masini, R.P., Medeiros, M.C. and Mendes, E.F. (2023). Machine learning advances for time series forecasting. Journal of Economic Surveys.
Goulet Coulombe, P., Leroux, M., Stevanovic, D. and Surprenant, S. (2022). How is machine learning useful for macroeconomic forecasting? Journal of Applied Econometrics.
Giannone, D., Lenza, M. and Primiceri, G.E. (2021). Economic predictions with big data: the illusion of sparsity. Econometrica.
Koop, G. and Korobilis, D. (2013). Large time-varying parameter VARs. Journal of Econometrics.
Gu, S., Kelly, B. and Xiu, D. (2020). Empirical asset pricing via machine learning. Review of Financial Studies.
Gentzkow, M., Kelly, B. and Taddy, M. (2019). Text as data. Journal of Economic Literature.
Kelly, B. and Xiu, D. (2023). Financial machine learning. NBER Working Paper 31502.

A comprehensive reading list will be provided to all enrolled participants.

Software / Hardware

Software:

Python run via Google Colab (requires a Google account)
Same code also available in MATLAB

Hardware:

Personal laptop capable of running statistical software
Minimum 8GB RAM recommended for computational exercises

Note

All necessary code examples and datasets will be provided
Participants will receive a free, time-limited MATLAB license before the program starts. Please install MATLAB on your computer before the course begins for use in practical sessions

Why join our Summer School?

All BSE Summer courses are taught to the same high standard as our Master’s programs. Join us to:

1

Network with like-minded peers

2

Study in vibrant Barcelona

3

Learn from world-renowned faculty

Admissions and Requirements

Participants must check they are eligible to take the course before applying.

Next edition: July 12-16, 2027

Early Bird deadline: April 15, 2027

Requirements

Summer School applicants normally demonstrate one or more of the following:

A strong background in Economics or a field closely related to the course topic (Statistics, Law, etc.)
Postgraduate degree or current Master’s/PhD studies related to the course topic
Relevant professional experience

Requirements for Forecasting with Machine Learning

Required knowledge:

Graduate-level (MSc) econometrics, including linear regression, maximum likelihood and time-series models
Familiarity with basic forecasting concepts and forecast evaluation
Basic Bayesian inference (priors, likelihoods and posteriors)
Working knowledge of linear algebra and probability

Technical skills:

Programming experience in Python, or MATLAB for laboratory sessions
Ability to work with economic datasets and perform statistical analysis

No prior background in machine learning is assumed.

Schedule

Here is your schedule for this edition of BSE Macroeconometrics Summer School Nowcasting and Forecasting course.

Time

12

mon

13

tue

14

wed

15

thu

16

fri

09:00-11:00

Lecture

14:30-16:00

Practical

Credit Transfers (ECTS)

To be eligible for credit transfer, students must complete a final project.

Students will deliver a short final project one week after the summer school finishes. It will consist in solving a final problem that will include the practical and empirical issues worked on in class.

Consult the Summer School Admissions page for more information about this option.

Certificate of Attendance

Participants who attend more than 80% of the course will receive a Certificate of Attendance, free of charge.

Course

Forecasting with Machine Learning

Advanced Bayesian Macroeconometrics

Bayesian Estimation of RANK and HANK Business Cycle Models

Introductory Bayesian Macroeconometrics

Introduction to Nowcasting and Forecasting

Structural Macroeconometrics: Foundations

Structural Macroeconometrics: Advanced Topics

Modality

Face-to-face

Total Hours

17.5

ECTS

1

Regular Fee

€1,470

Reduced Fee*

€840

* Reduced Fee applies for PhD or Master’s students, Alumni of BSE Master’s programs, and participants who are unemployed.

Multiple discounts are available.

For more information on discounts for participants funding their own course, please click here.

For more information on discounts for companies and institutions enrolling their employees, please click here.

Full discount conditions are available here.

FAQ

Here are some commonly asked questions by participants. Any further queries, please contact our Admissions Team.

Can I see the full Summer School calendar?

You can view the full Summer School calendar here.

Is accommodation included in the course fee?

Unfortunately, accommodation is not included in the course fee. Participants are responsible for finding accommodation.

Are the sessions recorded?

Sessions will NOT be recorded; however, the materials provided by the professor will be available for a month after the course has finished.

How much does each Summer School course cost?

Fees for each course may vary. Please consult each course page for accurate information.

Are there any discounts available?

Yes, BSE offers a variety of discounts on its Summer School courses. See more information about available discounts or request a personalized discount quote by email.

Can I take more than one course?

Yes! you can combine any of the Summer School courses (schedule permitting). See the full course calendar.

Cancelation and Refund Policy

Please consult BSE Summer School policies for more information.

Are there any evening activities during the course?

Yes, a social dinner is held once a week for all participants, it is free to attend.

Contact our Admissions Team

Mix and match your summer courses!

Remember that you can combine this program with courses in any of the other BSE Summer School programs (schedule permitting). Maximise your learning this summer and take advantage of our multiple-course discount.

Founding Institutions

Distinctions

Quality Transparency Legal Cookie Policy Contact Us Site map