Essays on risk, financial integration and learning

Rancoita, Elena

Preview

PDF
2015-05-18_Dissertation_Elena_Rancoita.pdf - Published
Download (1MB)

URL:	https://madoc.bib.uni-mannheim.de/38972
URN:	urn:nbn:de:bsz:180-madoc-389721
Document Type:	Doctoral dissertation
Year of publication:	2015
Place of publication:	Mannheim
University:	Universität Mannheim
Evaluator:	Adam, Klaus
Date of oral examination:	18 May 2015
Publication language:	English
Institution:	School of Law and Economics > VWL, Internat. Wirtschaftsbeziehungen (Adam)
Subject:	330 Economics
Subject headings (SWD):	Mehrgenerationenmodell , Risikoverteilung , Staatsbankrott , Staatsanleihe , Rentenmarkt , Adaptive Erwartung , Lernen , Zustandsraum , Nichtlineares System
Keywords (English):	Financial Integration , Default , Learning , Intregenerational Risk Sharing
Abstract:	This thesis consists of three self-contained chapters. The second and third chapters are theoretical works on two different inefficiencies of financial markets. In the second chapter, I show that in the most common Overlapping Generations models there is a tendency to under-invest in risky technologies due to the structural incompleteness of these models (the impossibility of trading with the unborn). In the third chapter, I ask if it is desirable to have large cross bond holdings within integrated financial markets or whether this leads to an excessive risk of default. The last chapter of this thesis, which is a joint work with Agustin Arias, is an empirical study on the estimation of adaptive learning models. We compare three alternative ways of estimating adaptive learning models and we show a new approach which can substantially reduce the computational costs of estimating those models, without impairing the quality of the estimates.