Bachelor/master dissertation (Dissertations and theses)
Análisis empírico del comercio internacional a partir de la segunda mitad del siglo XX.
Kozlowski, Diego
2019
 

Files


Full Text
tesis.pdf
Author postprint (52.7 MB)
Download

All documents in ORBilu are protected by a user license.

Send to



Details



Keywords :
World Trade; graphs; Data Science
Abstract :
[en] The study of international trade is one of the classical research areas of the economic thought. Its empirical characterization, at the same time that it goes back in time, is today an application space for new techniques, based on the intensive use of data. The increase in the availability of information curated for all countries allows exploring new analysis techniques. In turn, the multiple dimensions of the problem prevent a single analysis technique from capturing the entire phenomenon. The present work proposes to carry out various methodological proposals for the study of international trade, in such a way that together they allow obtaining a novel vision of the data. To do this, techniques widely used in other fields are tested on this particular subject, such as graph analysis and bayesian generative models. Aggregate trade between countries as well as the composition of the export baskets are considered.
Disciplines :
Computer science
Author, co-author :
Kozlowski, Diego ;  University of Luxembourg > Faculty of Science, Technology and Communication (FSTC) > Engineering Research Unit
Language :
Spanish
Title :
Análisis empírico del comercio internacional a partir de la segunda mitad del siglo XX.
Alternative titles :
[en] Empirical analysis of international trade from the second half of the twentieth century.
Defense date :
September 2019
Number of pages :
100
Institution :
University of Buenos Aires, Buenos Aires, Argentina
Degree :
Master in Data Mining and Knowledge Discovery
Promotor :
Semeshenko, Viktoriya
Aromí, Daniel
Ruz, Cecilia
Jury member :
Rosati, German
Focus Area :
Computational Sciences
Available on ORBilu :
since 15 July 2020

Statistics


Number of views
230 (11 by Unilu)
Number of downloads
114 (2 by Unilu)

Bibliography


Similar publications



Contact ORBilu