Submitted by Caroline Xavier (caroline.xavier@pucrs.br) on 2017-10-17T10:47:45Z No. of bitstreams: 1 DIS_CHRISTINY_KELLY_FERREIRA_NUNES_COMPLETO.pdf: 954134 bytes, checksum: 9b5f9a8ef9fc5629594411e21fc14351 (MD5) / Approved for entry into archive by Caroline Xavier (caroline.xavier@pucrs.br) on 2017-10-17T10:47:57Z (GMT) No. of bitstreams: 1 DIS_CHRISTINY_KELLY_FERREIRA_NUNES_COMPLETO.pdf: 954134 bytes, checksum: 9b5f9a8ef9fc5629594411e21fc14351 (MD5) / Made available in DSpace on 2017-10-17T10:48:05Z (GMT). No. of bitstreams: 1 DIS_CHRISTINY_KELLY_FERREIRA_NUNES_COMPLETO.pdf: 954134 bytes, checksum: 9b5f9a8ef9fc5629594411e21fc14351 (MD5) Previous issue date: 2017-08-04 / This paper seeks to determine the relationship between the default rate of companies and macroeconomic indicators. Analyzing the two base lines of credit from a Financial Institution in the period from 2012 to 2015. For this was parameterized a statistical model in which it contemplates idiosyncratic and macroeconomic variables, using logistic regression statistical technique. The objective of this paper is to capture the sensitivity of the probability of default of companies considering the analysis horizon of twelve months. The results show that the default is sensitivy to insertion of macroeconomic indicators, mainly in relation to PIB and PIB next year projection, named throughout the paper as PIB next. Despite the positive signs the results not shown so expressive like other papers in the area. It may have been impacted by a smaller base, if compared to the others. We consider the objective of this paper was achieved showing the viability, applicability of the model and their impacts. As well as that it can be used not only for data analysis, but also as an input into a portfolio credit risk model. / Esse trabalho buscou averiguar a rela??o entre a taxa de inadimpl?ncia de empresas e indicadores macroecon?micos, analisando a base de duas linhas de cr?dito de uma Institui??o Financeira no per?odo de 2012 a 2015. Para isso foi parametrizado um modelo estat?stico que contempla vari?veis idiossincr?ticas e vari?veis macroecon?micas, com t?cnica estat?stica de regress?o log?stica. A ideia foi capturar a sensibilidade da probabilidade de default das empresas considerando o horizonte de doze meses a partir da concess?o da linha de cr?dito. Os resultados mostraram que a inadimpl?ncia das linhas ? sens?vel ? inser??o de indicadores macroecon?micos, principalmente em rela??o ao PIB do ano e ao PIB projetado, denominado ao longo do trabalho como PIB pr?ximo. Apesar da sinaliza??o positiva, tais resultados, de forma geral, n?o se mostraram t?o expressivos quanto outros trabalhos da ?rea, podendo ter sido impactados por uma base menor, se comparados aos demais. A an?lise e dados e revis?o bibliogr?fica apontaram para a viabilidade e aplicabilidade do modelo e de seus impactos. Evidenciaram ainda que o modelo possa ser utilizado n?o somente para a an?lise de dados, mas tamb?m como insumo num modelo de risco de cr?dito de portf?lio.

Submitted by Sara Ribeiro (sara.ribeiro@ucb.br) on 2017-11-06T12:21:54Z No. of bitstreams: 1 DanielHenriqueSantanadePaivaDissertacao2016.pdf: 1413721 bytes, checksum: 479444b423ca48dd58e7b08c59d0bc38 (MD5) / Approved for entry into archive by Sara Ribeiro (sara.ribeiro@ucb.br) on 2017-11-06T12:22:09Z (GMT) No. of bitstreams: 1 DanielHenriqueSantanadePaivaDissertacao2016.pdf: 1413721 bytes, checksum: 479444b423ca48dd58e7b08c59d0bc38 (MD5) / Made available in DSpace on 2017-11-06T12:22:09Z (GMT). No. of bitstreams: 1 DanielHenriqueSantanadePaivaDissertacao2016.pdf: 1413721 bytes, checksum: 479444b423ca48dd58e7b08c59d0bc38 (MD5) Previous issue date: 2016-03-30 / The aim of this work is to evaluate whether risk variables and control directly affect banking activities, and how to affect them before the customers' financing need, and then allow credit expansion without increasing the exposure or the commitment of the National Financial System. For this we used the data model empirically estimated panel, suggested by Hsiao (1986). The sample covered data from a large financial institution in the country every month in 2015, from January to July to 2,093 business units. The main results suggest that risk and control variables affect significantly the "business", the "default" and the "result of the institution." In this line, should seek to identify new indications of operational risk present in the growth of economic activity and new sources of investment and financing with thirdparty funds (read financial institutions) to businesses. / O objetivo desta disserta????o ?? avaliar se vari??veis de risco e controle afetam diretamente as atividades banc??rias, e como as afetam, diante da necessidade de financiamento dos clientes, e ent??o possibilitar a expans??o de cr??dito sem elevar a exposi????o ou ao comprometimento do Sistema Financeiro Nacional. Para tanto foi utilizado o modelo de dados em painel estimado empiricamente, sugerido por Hsiao (1986). A amostra cobriu dados de uma grande institui????o financeira no pa??s, mensalmente no ano de 2015, de Janeiro a Julho, para 2.093 unidades de neg??cio. Os principais resultados sugerem que vari??veis de risco e controle afetam de forma sens??vel os ???neg??cios???, a ???inadimpl??ncia??? e o ???resultado da institui????o???. Nesta linha, deve-se buscar a identifica????o de novos ind??cios do risco operacional presente no crescimento da atividade econ??mica e novas fontes de investimentos e financiamentos com recursos de terceiros (leia-se Institui????es Financeiras) ??s empresas.

Submitted by Sara Ribeiro (sara.ribeiro@ucb.br) on 2017-04-20T12:41:30Z No. of bitstreams: 1 AugustoMarquesdeCastroOliveiraTese2016.pdf: 1480175 bytes, checksum: 685cbd36b095f75ce34ce81520f25b3e (MD5) / Approved for entry into archive by Sara Ribeiro (sara.ribeiro@ucb.br) on 2017-04-20T12:41:47Z (GMT) No. of bitstreams: 1 AugustoMarquesdeCastroOliveiraTese2016.pdf: 1480175 bytes, checksum: 685cbd36b095f75ce34ce81520f25b3e (MD5) / Made available in DSpace on 2017-04-20T12:41:47Z (GMT). No. of bitstreams: 1 AugustoMarquesdeCastroOliveiraTese2016.pdf: 1480175 bytes, checksum: 685cbd36b095f75ce34ce81520f25b3e (MD5) Previous issue date: 2016-04-05 / This work includes two studies. The first refers to an infinite horizon model of educational level choice, in which we find that in the long term, a stable relationship between wages and spending on education. On the demand side, it is assessed the impact of the workers` skill premium on the relative demand of qualified workers in the Brazilian Federal States (UF's). An analogous analysis if performed for the country members of the European Union (EU). The second study considers a dynamic model with habit formation and infinite horizon in which the consumer is a borrower which decides her level of consumption, payment of her debt and the new loans contract. It is estimated the parameters of this model using data from the Brazilian credit market and analyzed the sensitivity of the variables at steady state with respect to shocks in the interest rate to evaluate the impacts of monetary policy on the Brazilian credit market. / Este trabalho compreende dois estudos. O primeiro refere-se a um modelo de horizonte infinito de escolha de n??vel educacional, no qual encontramos que, no longo prazo, existe uma rela????o est??vel entre sal??rios e gastos em educa????o. Pelo lado da demanda, foi analisado o impacto do diferencial de sal??rios dos trabalhadores com diferentes n??veis de qualifica????o sobre a demanda das Unidades Federativas do Brasil (UF??s) e dos pa??ses integrantes da Uni??o Europeia (UE). O segundo estudo refere-se a um modelo din??mico com h??bitos no consumo e horizonte infinito no qual o consumidor ?? tomador de cr??dito e decide o seu n??vel de consumo, o pagamento de sua d??vida e a contrata????o de novos empr??stimos. Estimam-se os par??metros desse modelo utilizando os dados do mercado de cr??dito brasileiro e analisamos a sensibilidade das vari??veis no estado estacion??rio a choques na taxa de juros para verificar quais s??o os impactos da pol??tica monet??ria sobre o mercado de cr??dito brasileiro.