• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 1
  • Tagged with
  • 1
  • 1
  • 1
  • 1
  • 1
  • 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

Expectativas puras, prefer??ncia pela liquidez e modelos univariados "ARIMA" de Box & Jenkins projetam estruturas a termo de taxas de juros com efici??ncia?

Goulart, Lucio Allan 29 August 2005 (has links)
Made available in DSpace on 2015-12-03T18:32:43Z (GMT). No. of bitstreams: 1 Lucio_Allan_Goulart.pdf: 1745453 bytes, checksum: 18221402c2bb72ec072f698c188da21c (MD5) Previous issue date: 2005-08-29 / This study compared three ways of the yield to maturity curves behavior analysis in a specific negotiation time period of long term debt titles distributed at one determined period of time. Such mentioned debts are represented by two public issuers from different origins (Brazil and United States), with differentiated tenors in a Yield Time Structure (YTS), for each universe, denominated in a same currency, the United States Dollar. It was searched the definition of which methodology show greater forecasting capacity for the time data series at different time periods of YTS measurement. The time series analyses that indicate the yield behavior of the selected debts will be oriented on the principles of the Pure Expectations Theory (PET), the Liquidity Premium Theory (LPT) and the Univariate Box & Jenkins ARIMA Analysis. It was verified the low applicability of the use of PET and LPT for the YTSs in the two analyzed universes. For the ARIMA use, there is a reasonable acceptance for short term in the YTS of the United States, at the "knot" of measurement of 2 years. For the referring data to the YTS of Brazil, modeling ARIMA showed low forecasting capacity for all "knots" of the analyzed YTSs (2, 3, 4 and 5 years). For the tests of each proposed methodology, historical series of the active yields from the selected bonds had been analyzed, through the use of Excel spread sheets and the Minitab software analysis. The referring data to the historical series of the commented debts yields had been gotten at the Bloomberg L. P. electronic data system, with the previous authorization. / Este estudo comparou tr??s formas de an??lises comportamentais de curvas de juros, constitu??das pela taxa at?? o vencimento, ou yield to maturity, de t??tulos de d??vida de longo prazo, distribu??dos em um determinado per??odo de tempo. Tais curvas s??o resultantes de carteiras de t??tulos de dois perfis de emissores p??blicos de origem distinta (Brasil e Estados Unidos), com prazos diferenciados em uma Estrutura a Termo de Taxas de Juros (ETTJ), para cada universo. Todos os t??tulos s??o denominados na mesma moeda, o D??lar dos Estados Unidos. Foi buscada a defini????o da metodologia que apresentasse maior capacidade de previs??o para as s??ries temporais de dados que constitu??ssem uma ETTJ em momentos diferenciados. A an??lise de s??ries temporais que melhor retratassem o comportamento da taxa ativa das ETTJ foi feita com base na Teoria das Expectativas Puras (TEP), Teoria de Prefer??ncia pela Liquidez (TPL) e An??lise Univariada ARIMA de Box & Jenkins. Houve a verifica????o da baixa aplicabilidade do uso de TEP e TPL para as ETTJ dos dois universos analisados. Para o uso de ARIMA, houve uma aceita????o razo??vel para o curto prazo na ETTJ dos Estados Unidos, no "n??" de medi????o de 2 anos. Para os dados referentes ?? ETTJ do Brasil, a modelagem ARIMA mostrou pouca previsibilidade para todos os "n??s" das ETTJ analisadas (2, 3, 4 e 5 anos). Para os testes de cada metodologia proposta foram analisadas s??ries hist??ricas das taxas ativas dos t??tulos selecionados, atrav??s do uso de planilhas ExcelTM e por an??lise atrav??s do software MinitabTM. Os dados referentes ??s s??ries hist??ricas das taxas ativas dos t??tulos de d??vida comentados foram obtidos no sistema Bloomberg L. P. de informa????es eletr??nicas, com a devida autoriza????o.

Page generated in 0.1325 seconds