Marcela Galvis-Restrepo

Marcela Galvis-Restrepo
Copenhagen Business School · Department of Economics

About

11
Publications
1,724
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
35
Citations
Citations since 2016
8 Research Items
31 Citations
201620172018201920202021202201234567
201620172018201920202021202201234567
201620172018201920202021202201234567
201620172018201920202021202201234567
Additional affiliations
August 2018 - present
Copenhagen Business School
Position
  • PhD Student
Education
August 2018 - June 2022
Copenhagen Business School
Field of study
  • Machine learning
September 2012 - September 2013
Friedrich Schiller University Jena
Field of study
  • Economics of innovation
January 2004 - September 2009
University of Antioquia
Field of study
  • Economics

Publications

Publications (11)
Preprint
Full-text available
In many education systems around the world, the problem of school dropout is widespread. Identifying students at a higher risk of dropping out is crucial in developing settings, where the cost of preventive interventions is a major concern. By using individual-level administrative data on school enrollment in rural Antioquia, Colombia, I apply supe...
Preprint
Full-text available
In recent years, supervised classification has been used to support or even replace human decisions in high stakes domains. The training of these algorithms uses historical data which might be biased against individuals with certain sensitive attributes. The increasing concern about potential biases has motivated anti-discrimination laws prohibitin...
Preprint
Full-text available
We propose a method to reduce the complexity of Generalized Linear Models in the presence of categorical predictors. The traditional one-hot encoding, where each category is represented by a dummy variable, can be wasteful, difficult to interpret, and prone to overfitting, especially when dealing with high-cardinality categorical predictors. This p...
Article
Full-text available
We propose a method to reduce the complexity of Generalized Linear Models in the presence of categorical predictors. The traditional one-hot encoding, where each category is represented by a dummy variable, can be wasteful, difficult to interpret, and prone to overfitting, especially when dealing with high-cardinality categorical predictors. This p...
Preprint
Full-text available
In this paper, our goal is to enhance the interpretability of Generalized Linear Models by identifying the most relevant interactions between categorical predictors. In the presence of categorical predictors, searching for interaction effects can quickly become a highly combinatorial problem when we have many categorical predictors or even a few, b...
Preprint
Full-text available
We propose a method to reduce the complexity of Generalized Linear Models in the presence of categorical predictors. The traditional one-hot encoding, where each category is represented by a dummy variable, can be wasteful, difficult to interpret, and prone to overfitting, especially when dealing with high-cardinality categorical predictors. This p...
Poster
Research was institutionalized in Colombian universities around fifty years ago with the creation of government institutions to stimulate this activity, and the support of loans from the Inter-American Development Bank –IDB- that were used to start building scientific and technological capabilities (Bucheli et al., 2012; CINDA, 2012; OECD & The Wor...
Thesis
Full-text available
Se analiza la formación de coaliciones regionales durante una de las fases de la votación del Plan Nacional de Inversiones Públicas con ayuda del concepto de pork barrel; y se realiza una propuesta metodológica para estudiar dichas coaliciones con la teoría de juegos simples, utilizando los índices de Shapley-Shubick y Banzhaf en la determinación c...

Network

Cited By

Projects

Projects (2)
Project
NeEDS (Network of European Data Scientists) provides an integrated modelling and computing environment that facilitates data analysis and data visualization to enhance interaction. NeEDS brings together an excellent interdisciplinary research team that integrates expertise from three relevant academic disciplines, Mathematical Optimization, Visualization and Network Science, and is excellently placed to tackle the challenges. NeEDS develops mathematical models, yielding results which are interpretable, easy-to-visualize, and flexible enough to incorporate user knowledge from complex data. These models require the numerical resolution of computationally demanding Mixed Integer Nonlinear Programming formulations, and for this purpose NeEDS develops innovative mathematical optimization based heuristics.