XREAP2017-04: Forecasting compositional risk allocations

We analyse models for panel data that arise in risk allocation problems,when a given set of sources are the cause of an aggregate risk value. We focus on the modeling and forecasting of proportional contributions to risk. Compositional data methods are proposed and the regression is flexible to incorporate external information from other variables. We guarantee that projected proportional contributions add up to 100%, and we introduce a method to generate confidence regions with the same restriction. An illustration using data from the stock exchange is provided.

Boonen, T.J.; Guillén, M. (Riskcenter, XREAP); Santolino, M. (Riskcenter, XREAP)


XREAP2017-03: What drives migration moves across urban areas in Spain? Evidence from the Great Recession

In Spain, economic disparities between regions have traditionally played a relevant role in migration. Nevertheless, during the previous high-instability period, analyses provided conflicting results about the effect of these variables. In this work, we aim to determine the role that labour market factors play in internal migration during the Great Recession, paying special attention to the migration response of the heterogeneous population groups. To do so, we resort to an extended gravity model and we consider as a territorial unit the 45 Spanish Functional Urban Areas. Our results point to real wages as having a significant influence on migration motivations.

Melguizo, C. (AQR-IREA, XREAP); Royuela, V. (AQR-IREA, XREAP)



This paper analyzes the drivers of wage differences among college graduates who hold a degree in a different field of study. We focus on Turkey, an emerging country that is characterized by a sustained expansion of higher education. We estimate conditional wage gaps by field of study using OLS regressions. Average differentials are subsequently decomposed into the contribution of observable characteristics (endowment) and unobservable characteristics (returns). To shed light on distributional wage disparities by field of study, we provide estimates along the unconditional wage distribution by means of RIF-Regressions. Finally, we also decompose the contribution of explained and unexplained factors in accounting for wage gaps along the whole distribution. As such, this is the first work providing evidence on distributional wage differences by college major for a developing country. The results indicate the existence of important wage differences by field of study, which are partly accounted by differences in observable characteristics (especially occupation and, to a lesser extent, employment sector). These pay gaps are also heterogeneous over the unconditional distribution of wages, as is the share of wage differentials that can be attributed to differences in observable characteristics across workers with degrees in different fields of study.

Di Paolo, A. (AQR-IREA, XREAP); Tansel, A.



In this paper, we examine the impact of non-stop flights on the connectivity of European cities with distant locations from the rest of the world. We use data on inter-city passenger flows including non-stop and connecting traffic so that we have a precise measure of the economic and social links between cities. We apply a matching procedure and run regressions using instrumental variables to deal with the potential endogeneity bias of the variables for non-stop flights. We find a strong causal relationship between the amount of total traffic and the supply of non-stop long-haul flights in the considered inter-city markets. Traffic increases from the shift from ‘not having’ to ‘having’ non-stop flights can be more than double. Such increase in the amount of traffic does not seem to be related with a systematic change in fares.

Bernardo, V.  (GiM-IREA, XREAP); Fageda, X. (GiM-IREA, XREAP)