Prediction and learning of exporting firms: a study of Colombia

Simple item page
dc.contributor.MentorHallak, Juan Carlos
dc.contributor.MentorSosa Escudero, Walter
dc.creator.AutorMannarino, Valentín
dc.date.accessioned2026-04-08T18:13:24Z
dc.date.available2026-04-08T18:13:24Z
dc.date.issued2025-12
dc.descriptionFil: Mannarino, Valentín. Universidad de San Andrés. Departamento de Economía; Argentina.
dc.description.abstractThis thesis applies machine learning techniques to predict which manufacturing firms in Colombia are likely to become exporters, using data from the Encuesta Anual Manufacturera (EAM) and Encuesta de Desarrollo e Innovaci´on Tecnol´ogica (EDIT) for the period 2015–2019. The objective is to estimate each firm’s “distance to export” through a probability score learned from the characteristics of existing exporters. Among the different algorithms tested, Logit with LASSO regularization delivers the best predictive performance, correctly identifying nearly three out of four actual exporters. Building on these predictions, the study introduces an exporting score, a probability measure that ranks firms by their proximity to the export margin. This score captures heterogeneity among non-exporters, anticipates entry and exit dynamics, and highlights sectoral and geographic clusters of latent export potential. In addition, the analysis shows that a set of firm level characteristics consistently emerge as the most relevant predictors across models: importer status, firm size, and combined spillovers, complemented by operational variables such as value added, inventories, and quality certification. The exporting score also reveals a transition zone around a score of 0.55–0.58, which delineates the range where policy support can be most effective, either by activating firms close to exporting or by preventing the exit of current exporters. Beyond its analytical value, the score provides a practical input for policy design and evaluation, allowing export promotion agencies to target resources more efficiently, define eligibility thresholds, and even implement randomized or regression-discontinuity designs. These findings highlight the potential of predictive approaches to enhance export promotion under budget constraints, maximizing the impact of limited public resources on export growth and retention.
dc.formatapplication/pdf
dc.identifier.urihttps://repositorio.udesa.edu.ar/handle/10908/26296
dc.languageeng
dc.publisherUniversidad de San Andrés. Departamento de Economía
dc.rightsinfo:eu-repo/semantics/openAccess
dc.rightshttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.titlePrediction and learning of exporting firms: a study of Colombia
dc.typeTesis
dc.typeinfo:eu-repo/semantics/masterThesis
dc.typeinfo:ar-repo/semantics/tesis de maestría
dc.typeinfo:eu-repo/semantics/updatedVersion
Files
Original bundle
Thumbnail Image
Name:
[P][W] T. M. Eco. Mannarino, Valentín.pdf
Size:
7.19 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Tesis de Maestría en Economía