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

[en] GROWING FUNCTIONS: AN OVERVIEW / [pt] CURVAS DE CRESCIMENTO: UMA VISÃO GERAL

LEANDRO SAUER 16 September 2009 (has links)
[pt] Muitas vezes são requeridas previsões a médio e longo prazo, de um determinado fenômeno, sem ter conhecimento específico de que tipos de fatores estão, ou estarão, influenciando-o. Por isso soa usadas funções matemáticas para modelar o respectivo fenômeno. Apresentadas um tratamento para fenômenos monotonicamente crescentes com um limite assintótico. As funções matemáticas usadas são chamadas curvas de crescimento de forma 8, dentre estas trabalhamos com a família exponencial modificadas generalizada. Apresentamos uma abordagem estática, dinâmica clássica e dinâmica Bayesiana com fatores de desconto e aplicamos a metodologia exposta a seis séries reais: Número de notificações de casos de AIDS, porcentagem de domicílios com telefone nos EUA, porcentagem de domicílios com sistema eletrônico móvel para telefone nos EUA, porcentagem de tecido manufaturado nos EUA, estoque de tratores na Espanha e porcentagem de domicílios com TV preto e branco nos EUA. / [en] It is not uncommon to find situations in which on is asked to make both lony and short term predictions for a particular phenomena, without having at this or her disposal the necessary information concerning the factors which might be directly affecting the phenomena ins question. Given this state of affairs, the use of mathematical functions in modelling becomes justified. Our aim in this thesis in to model monotonically growing phenomena which present an asymptotic limit. The mathematical functions entertained here are those know as growing functions with as S shape . In particular 8 shape. In particular we focus our attention on those function from the generalized modified exponential family. The formulations we exploit are the statistic, the dynamic classic and the Bayesian dynamic with discount factors. We apply the aforementioned methodologies to six time series: the number of AIDS cases in Brazil, the percentage of USA homes with a telephone, the percentage of USA homes with a wirelesss telephone, the percentage of fabrics manufactured in the USA as compared to the total USA market, the stock of tractors in Spain and the percentage of USA homes with a monochromatic TV set.

Page generated in 0.0243 seconds