Doria, Omar (2016) Essays on the Use of Big Data in Development Economics. Advisor: Pammolli, Prof. Fabio. pp. 147. [IMT PhD Thesis]
|
Text
Doria_phdthesis.pdf - Published Version Available under License Creative Commons Attribution Non-commercial Share Alike. Download (10MB) | Preview |
Abstract
In this thesis I approach problems within the literature of Development Economics. Using tools from policy evaluation, different quantitative methods and big data sources, I study the common problems that affect the development of the nations. I separate my thesis into three chapters. In Chapter 1, using policy evaluation techniques together with other quantitative methods, I study the effects of the policy integration for the academic sector within the European Union. In the second chapter, I study one of the most important subjects presented in this thesis: inequality. Using the case of Colombian municipalities, I examine how international trade affects social conditions measured by the Multidimensional Poverty Index. Finally, in the third chapter, I study the effect of the patent innovation using the complexity algorithm developed by Hidalgo and Haussmann. Here I do a comparative of the patent innovation using two aggregations: countries and cities. Next, I will explain in more detail the findings of each chapter. Chapter 1: It is generally accepted that the frequency of cross-border collaborations has been increasing in recent decades, which is principally regarded as a symptom of globalization. While this is true on average, we uncover a more nuanced story by analyzing publications, and by disaggregating these R&D outputs by country, across 14 well-defined research subject areas. In this way, we are able to interpret trends in cross-border activities according to more domain-specific trends. We focus our analysis on the impact of entry into the EU by new member states by quantifying the rate of cross-border collaboration before and after the 2004 enlargement of the European Union. In this sense, we build upon recent studies aimed at quantifying the impact of European Research Area integration policies on the activity of the European innovation system. We combine descriptive complex networks techniques with panel regression (Difference in Difference) methods to reveal, counterintuitively, a decrease in cross-border activity by the new EU member states following their entrance. The results show that while the number of crossborder collaborations in academia is increasing in the old member countries, and despite that the number of cross-border publications in the new members is higher compared with past years, they would actually collaborate more being outside the European Union. We use data for the inventor mobility network to show that these counterintuitive trends are none other than the negative externalities of unification associated with brain-drain. Chapter 2: We empirically measure the effects of international trade on inequality. By studying the Colombian case, we found that the municipalities with exporter firms have an 11% greater probability for increasing their inequality through social deprivations compared to municipalities, where any of the firms are exporters. Furthermore, we aggregate firms’ exports at the municipality level by using the minimum economical unit affected by the incoming wealth from foreign markets, considering spatial relations to account for direct, indirect and total effects. We define social inequality as the average shortfall of social conditions by municipality. Specifically we use the Colombian Multidimensional Poverty Index. As a result, we found empirical evidence for a strong neighborhood effect, which helps make the decision that would be used to improve social conditions in those municipalities without exporter firms. Chapter 3: One of the most important questions in Developmental Economics is how technological innovation is able to shift development. Here, we use the Hidalgo and Hausmann complexity algorithm to estimate how the selection of the innovation field affects the leadership in innovation among countries by using the first patent of triadic families of the European, Japanese and United States patent office. In this analysis we rank countries, regions and patents using the Economic Complexity Index (ECI) and Product Complexity Index (PCI). Our findings highlight the United States as a leading country in patent innovation during most of the years. In contrast, using the region aggregation level, we find that the Japanese regions are the leaders in patent innovation during every year in our data. However, the most complex regions in the Unites States. On the patent side, we note that the fields related to chemistry, biotechnology and pharmaceuticals play a very important role in patent innovation. Finally, we compare our findings with similar works of other researchers, finding a strong relationship between academic research and patent innovation. In this thesis, I explore quantitative methods and Data Science techniques applied to social studies for different aggregations. In the first chapter, I account for the collaboration in R&D between groups of countries. In the second chapter, I explore how one single country and its municipalities relate to the world through trade, and how its relationship could modify the condition, while its minimal political level affects the social conditions through a mechanism accepted and studied by classical economics. In Chapter 3, we study the behavior of nations, where every nation with innovative production is included. By alternating the level of aggregations from a nation to a city, we are able to determine the role that political regions play within the whole country. Recognizing the importance of the results and the methods that I use to explore these important development issues, such as international integration, inequality, international trade, innovation and relation country-cities, this thesis questions the classical economical methods and the relevance in accepting new methodologies based on Data Science to answer the question of classical problems that were answered by using theoretical methods in the past. These new methodologies based on Big Data are evolving every day with importance, and, moreover, with information collected by governments, social networks, international organizations and private institutions. Therefore, the methods exposed in this dissertation play a fundamental role in the reshaping of economics and the study and interpretation of the relation between governments, societies and citizens.
| Item Type: | IMT PhD Thesis |
|---|---|
| Subjects: | H Social Sciences > HB Economic Theory |
| PhD Course: | Economics, Markets, Institutions |
| Identification Number: | https://doi.org/10.6092/imtlucca/e-theses/207 |
| NBN Number: | urn:nbn:it:imtlucca-27235 |
| Date Deposited: | 22 Mar 2017 14:12 |
| URI: | http://e-theses.imtlucca.it/id/eprint/207 |
Actions (login required, only for staff repository)
 |
View Item |
