How to deal with noise

The communication channels we use in our daily lives (phone calls, TV broadcasts, the Internet...) all have inherent noise, although we barely notice it because of underlying error-correcting procedures. In this course we will cover the basics of the mathematics of reliable communication, which go back to groundbreaking work of Hamming and Shannon in the mid-20th century that kickstarted the fields of coding theory and information theory.

João is an assistant professor in the Department of Mathematics at Técnico-ULisboa and a researcher at Instituto de Telecomunicações. Previously, he was an assistant professor in the CS Department of Universidade Nova de Lisboa. Before that, he was a post doctoral fellow in the CS Department of Carnegie Mellon University, hosted jointly by Vipul Goyal and Venkatesan Guruswami. João received his PhD from the Department of Computing of Imperial College London, where he was advised by Mahdi Cheraghchi. Before that, he received an MSc in Computer Science from ETH Zurich and a BSc in Applied Mathematics and Computation from Técnico-ULisboa. He likes thinking about coding theory, information theory, and the theory of computation, and in 2025 was awarded an ERC Starting Grant.

Spurious Statistical Inference in Time-Series Analysis and Dynamic Modeling

This minicourse offers a primer on the detection and avoidance of spurious statistical inference in time-series analysis and dynamic modeling. We examine four sources of spurious inference that frequently arise in empirical practice. These include spurious regression in the presence of non-stationary time series; misleading conclusions arising from the conflation of short-term and long-term dynamic relationships; inference distortions induced by model misspecification; and spurious causal inference, where purely predictive models are mistakenly interpreted as causal. For each of these issues, we discuss both the underlying mechanisms that produce spurious results and the methodological solutions available to practitioners.

Francisco Blasques is Professor of Econometrics and Data Science at Vrije Universiteit Amsterdam and Invited Professor at Instituto Superior Técnico, Universidade de Lisboa, and Fellow of the Tinbergen Institute. His research focuses on statistical inference, filtering, dynamic modeling, and time-series analysis. He has led R&D projects for governmental institutions, central banks and private companies worldwide, and leads the development of software applications in AI, data science, and statistical modeling. He has served as Director of Research at Vrije Universiteit Amsterdam, and as  Coordinator and Director of the Bachelor's and Master's programs in Data Science, Econometrics and Operations Research.