Thesis Análisis de correlaciones canónicas entre las variables de capital social, innovación y macreoeconomía, en los países de América.
Loading...
Date
2018-11-14
Authors
Journal Title
Journal ISSN
Volume Title
Program
Ingeniería Civil Industrial
Departament
Campus
Campus Santiago Vitacura
Abstract
Esta investigación aborda la relación entre los grupos de capital social, innovación y variables macroeconómicas en los países de América. Se analizan datos de PNUD, universidades y diferentes organizaciones relacionadas.
La metodología del diagrama de caja se aplica a cada variable para detectar observaciones atípicas con el fin de excluir a los países que no son adecuados para cumplir los objetivos, o que conducen a errores en la interpretación de los resultados. Asimismo, para evitar problemas de multicolinealidad, continuamos con la construcción de gráficos y el análisis de correlaciones entre las variables de cada uno de los grupos para asegurar que los indicadores que representan el Capital Social, la Innovación y las variables Macroeconómicas, no estén correlacionados entre sí mismos.
Por último, una vez que los países y las variables se seleccionan para su inclusión en el proyecto, se puede aplicar un Análisis canónico para determinar nuevas variables canónicas y mejorar la eficiencia del análisis.
Se otorga reconocimiento al proyecto interno No. PIL1713 de la Universidad Técnica Federico Santa María por apoyar este trabajo.
This research deals with the relationship between the groups of Social Capital, Innovation and Macroeconomic variables in the countries of America. Data from UNDP, Universities and different related organizations is analyzed. Box plot methodology is applied to each variable to detect atypical observations in order to leave out countries that are not adequate for meeting the objectives, or that lead to errors in the interpretation of the results. Likewise, to avoid problems of multicollinearity, we continue with the construction of graphs and analysis of correlations between the variables of each of the groups in order to ensure that the indicators that represent Social Capital, Innovation and variables Macroeconomic, are not correlated in between among themselves. Lastly, once the countries and variables are selected for inclusion into the project, a Canonical Analysis can be applied to determine new canonical variates and so improve the efficiency of the analysis. Acknowledgement is given to the Universidad Técnica Federico Santa María internal project No. PIL1713 for supporting this work.
This research deals with the relationship between the groups of Social Capital, Innovation and Macroeconomic variables in the countries of America. Data from UNDP, Universities and different related organizations is analyzed. Box plot methodology is applied to each variable to detect atypical observations in order to leave out countries that are not adequate for meeting the objectives, or that lead to errors in the interpretation of the results. Likewise, to avoid problems of multicollinearity, we continue with the construction of graphs and analysis of correlations between the variables of each of the groups in order to ensure that the indicators that represent Social Capital, Innovation and variables Macroeconomic, are not correlated in between among themselves. Lastly, once the countries and variables are selected for inclusion into the project, a Canonical Analysis can be applied to determine new canonical variates and so improve the efficiency of the analysis. Acknowledgement is given to the Universidad Técnica Federico Santa María internal project No. PIL1713 for supporting this work.
Description
Keywords
Correlación canónica, Modelos estadísticos, Métodos cuantitativos, Macroeconomía
