Estada del 9 de gener a l'11 de març de 2017

Title: Modeling rational bubbles and capital intermediation using compositional data analysis

**I. ****Modeling rational bubbles**

In this part, I will analyze the correlation between rational bubbles in the exchange rate market in a group of countries. Specifically, for each country in a group, I will define a fundamental value for the exchange rate and the deviations of the spot exchange rate from its fundamental value will be the bubble of it in the country. Finally, I will analyze the joint dynamics of the exchange rates of the group of countries.

In addition, a new concept of exchange rate bubble will be defined. Taking a compositional dynamical system of all the spot exchange rates in the group, the steady state of that dynamical system will define the fundamental value for each country; therefore, any temporary deviation from those values will be the compositional exchange rate bubbles

**II. ****Capital intermediation**

The Property Income is the rental paid to the owners of financial assets, or non-produced assets, in return for its use. Examples of those rentals are: interest paid by debts, dividends distributed by the issuers of certain stocks, returns of the investment made by insurance companies of technical reserves and land rentals. All those rentals have an associated stock and a rate of return. Using data from the National Accounts of Brazil from 1995 to 2013, I will calibrate one of those variables (stocks or rates of return). After that, I will analyze the compositional behavior of both and the stability of the long run behavior of the relative amounts (rentals, stocks and rates of return). As a result I expect to understand the co-movements of the rentals and if some institution present more concentration in capital intermediation. This important issue will shed light how the income circulates in the economy using the financial intermediation.

Proponent: Wilfredo Fernando Leiva Maldonado

https://sites.google.com/site/wilfredoflmaldonado/

]]>

This CoDaCourse provides an introduction to theoretical and practical aspects of the statistical analysis of compositional data, as well as an informal discussion forum on more advances topics. In the theoretical sessions, the current state of the art in this field will be presented. In the practical sessions, coda techniques are applied using the available software CoDaPack and some R packages. An open discussion session is also planned. We encourage you to bring your own data!!!

Statisticians and applied scientists of any field, in particular engineering geologists, geologists, economics, scientists, environmental engineers, and bio&environmental scientists, working in academic or industrial institutions, are strongly encouraged to take this CoDaCourse. It is recommended that attendants have undergone some first semester courses on statistics. Basic knowledge about multivariate statistics may also be handy.

Due to the class room facilities, course seats are limited to 30. Early registration is recommended. Course may get cancelled if registered students are less than 5.

To apply and for more information visit the website: http://www.compositionaldata.com/codacourses.php

This course is organized by the Research Group on Compositional Data Analysis from the Computer Science, Applied Mathematics and Statistics Department of the University of Girona (UdG). The course is officially accredited by the International Association for Mathematical Geosciences (IAMG).]]>