• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 1
  • Tagged with
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Aplicação de redes neurais na precificação de debêntures

Curi, Leonardo Zago 07 February 2008 (has links)
Made available in DSpace on 2010-04-20T21:00:06Z (GMT). No. of bitstreams: 3 leonardozago.pdf.jpg: 15861 bytes, checksum: c8779d38fde5b427b0a2bb1776baec3d (MD5) leonardozago.pdf.txt: 73279 bytes, checksum: b941e1925fefbef1bd86777bd72dccd5 (MD5) leonardozago.pdf: 383709 bytes, checksum: a91da81e0abe11c9497e8a77266f42fa (MD5) Previous issue date: 2008-02-07T00:00:00Z / Previous studies on pricing of Corporate Bonds have shown that prices for these securities in Brazil cannot be explained only by credit risk, but also by other factors, such as liquidity risk. On the other hand, other studies also have shown that neural networks models have been more successful than traditional models in explaining issues related to corporate bonds, such as modeling default probabilities and ratings from agencies such as Standard & Poors and Moodys. The purpose of this study is to test neural networks technique in pricing corporate bonds in Brasil and compare the results obtained with the ones obtained through linear regressions. To accomplish this, accounting variables and specific features of a bond such as time to maturity and calllable features were used as independent variables. Regarding dependent variables, ANDIMA’s daily rates were used as a reference for market value for corporate bonds. The variables described above were tested in several models through ordinary least squares and the model which presented the best result was also tested in neural networks with two hidden layers. The neural networks with six and eight neurons presented better results than models estimated through pooling and ordinary least squares both in the training stage as in the testing one. Nonetheless, there’s still much room for improvement in the models considering the size of the database available is still small and the rates published by ANDIMA are averages of a small group of financial institutions and may not reflect the true market value of a corporate bond. / Estudos anteriores mostraram que a técnica de redes neurais tem sido mais bem sucedida que os modelos tradicionais em vários assuntos relacionados ao mercado de debêntures, tais como modelar a probabilidade de default e em explicar os ratings de agências classificadoras de risco, como Standard & Poors e Moodys. O objetivo deste trabalho é testar a técnica de redes neurais para precificar debêntures no Brasil e comparar os resultados obtidos com regressões lineares. Para isso, utilizaram-se como variáveis explicativas dados contábeis, características específicas das emissões, tais como prazo para vencimento e cláusulas de recompra antecipada. Em relação às variáveis dependentes, optou-se por utilizar as taxas divulgadas diariamente pela ANDIMA como valor de mercado para as debêntures. As variáveis acima foram testadas em diversos modelos pelo método dos mínimos quadrados ordinários e o modelo que apresentou o melhor resultado foi testado em redes neurais com duas camadas intermediárias. Os resultados obtidos com redes neurais com seis e oito neurônios apresentaram resultados superiores aos modelos estimados por mínimos quadrados ordinários tanto na fase de treinamento como na fase de testes. No entanto, ainda há bastante espaço para melhorias nos modelos dado que o tamanho da base de dados disponível para este tipo de testes no Brasil ainda não é a ideal e as taxas divulgadas pela ANDIMA são médias de um grupo pequeno de instituições e não necessariamente refletem o valor de mercado de uma debênture.

Page generated in 0.0578 seconds