Submitted by Roberta Lorenzon (roberta.lorenzon@fgv.br) on 2011-05-30T14:26:15Z
No. of bitstreams: 1
63080100015.pdf: 781749 bytes, checksum: 32dd660d87d63789ab4a0ee0600ce50d (MD5) / Approved for entry into archive by Gisele Isaura Hannickel(gisele.hannickel@fgv.br) on 2011-05-30T14:47:36Z (GMT) No. of bitstreams: 1
63080100015.pdf: 781749 bytes, checksum: 32dd660d87d63789ab4a0ee0600ce50d (MD5) / Approved for entry into archive by Gisele Isaura Hannickel(gisele.hannickel@fgv.br) on 2011-05-30T14:48:59Z (GMT) No. of bitstreams: 1
63080100015.pdf: 781749 bytes, checksum: 32dd660d87d63789ab4a0ee0600ce50d (MD5) / Made available in DSpace on 2011-05-31T15:39:00Z (GMT). No. of bitstreams: 1
63080100015.pdf: 781749 bytes, checksum: 32dd660d87d63789ab4a0ee0600ce50d (MD5)
Previous issue date: 2010-05-28 / The present work aims to study the macroeconomic factors influence in credit risk for installment autoloans operations. The study is based on 4.887 credit operations surveyed in the Credit Risk Information System (SCR) hold by the Brazilian Central Bank. Using Survival Analysis applied to interval censured data, we achieved a model to estimate the hazard function and we propose a method for calculating the probability of default in a twelve month period. Our results indicate a strong time dependence for the hazard function by a polynomial approximation in all estimated models. The model with the best Akaike Information Criteria estimate a positive effect of 0,07% for males over de basic hazard function, and 0,011% for the increasing of ten base points on the operation annual interest rate, toward, for each R$ 1.000,00 on the installment, the hazard function suffer a negative effect of 0,28% , and an estimated elevation of 0,0069% for the same amount added to operation contracted value. For de macroeconomics factors, we find statistically significant effects for the unemployment rate (-0,12%) , for the one lag of the unemployment rate (0,12%), for the first difference of the industrial product index(-0,008%), for one lag of inflation rate (-0,13%) and for the exchange rate (-0,23%). We do not find statistic significant results for all other tested variables. / The present work aims to study the macroeconomic factors influence in credit risk for installment autoloans operations. The study is based on 4.887 credit operations surveyed in the Credit Risk Information System (SCR) hold by the Brazilian Central Bank. Using Survival Analysis applied to interval censured data, we achieved a model to estimate the hazard function and we propose a method for calculating the probability of default in a twelve month period. Our results indicate a strong time dependence for the hazard function by a polynomial approximation in all estimated models. The model with the best Akaike Information Criteria estimate a positive effect of 0,07% for males over de basic hazard function, and 0,011% for the increasing of ten base points on the operation annual interest rate, toward, for each R$ 1.000,00 on the installment, the hazard function suffer a negative effect of 0,28% , and an estimated elevation of 0,0069% for the same amount added to operation contracted value. For de macroeconomics factors, we find statistically significant effects for the unemployment rate (-0,12%) , for the one lag of the unemployment rate (0,12%), for the first difference of the industrial product index(-0,008%), for one lag of inflation rate (-0,13%) and for the exchange rate (-0,23%). We do not find statistic significant results for all other tested variables.
| Identifer | oai:union.ndltd.org:IBICT/oai:bibliotecadigital.fgv.br:10438/8267 |
| Date | 28 May 2010 |
| Creators | Andrade, Rodrigo Augusto Silva de |
| Contributors | Escolas::EESP, Picchetti, Paulo |
| Source Sets | IBICT Brazilian ETDs |
| Language | Portuguese |
| Detected Language | English |
| Type | info:eu-repo/semantics/publishedVersion, info:eu-repo/semantics/masterThesis |
| Source | reponame:Repositório Institucional do FGV, instname:Fundação Getulio Vargas, instacron:FGV |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0025 seconds