z-logo
open-access-imgOpen Access
PROPOSTA METODOLÓGICA PARA O AJUSTE ÓTIMO DA DISTRIBUIÇÃO DIAMÉTRICA WEIBULL 3P
Author(s) -
Oscar Santiago Vallejos Barra,
Carlos Roberto Sanquetta,
Júlio Eduardo Arce,
Sebastião do Amaral Machado,
Ana Paula Dalla Côrte
Publication year - 2004
Publication title -
floresta
Language(s) - English
Resource type - Journals
eISSN - 1982-4688
pISSN - 0015-3826
DOI - 10.5380/rf.v34i3.2425
Subject(s) - weibull distribution , mathematics , statistics , physics
A distribuicao de Weibull de tres parâmetros tem ampla utilizacao na area florestal. Existem tres metodos para ajustar a distribuicao, os quais consideram o parâmetro de locacao como um termo independente que deve ser conhecido para obter os restantes parâmetros. Esta proposta metodologica visa, atraves de um processo interativo, otimizar o ajuste de cada um dos metodos mais utilizados para esta finalidade, sendo eles: maxima verossimilhanca, momentos e percentis. Esta proposta visa minimizar o “dn” do teste de aderencia de Kolmogorov-Smirnov. Observou-se que os valores de “dn” da distribuicao Weibull de tres parâmetros sao inferiores aos obtidos na de dois parâmetros nos tres metodos de ajuste. Observou-se ainda que os valores “dn” de cada metodo nao apresentam diferencas expressivas, mas quando sao comparadas as probabilidades associadas a magnitude tornam-se relevantes e justificam a metodologia proposta. Concluiu-se que esta nova metodologia e uma alternativa util para ajuste de distribuicoes diametricas e aplicacoes em modelagem do crescimento e da producao de povoamentos florestais. A NEW METHOD FOR THE OPTIMUM FITTING OF THE 3-P WEIBULL DIAMETER DISTRIBUTION Abstract The Weibull probability distribution of three parameters has wide use forestry. There are three fitting methods used for this purpose, which take into consideration the location parameter as an independent term that should be known previously to obtain the remaining parameters. This methodological proposal aims at showing an iterative method of optimization the fitting of the three parameters of the Weibull function in each one of the methods: maximum likelihood, moments, and percentiles. The proposed method minimizes the statistical “dn” of the Kolmogorov-Smirnov test of goodness-of-fit. It was noticed that “dn” of the three parameters Weibull distribution are lower than those of the two parameters function for the three fitting methods. It was also observed that “dn” values of each method were not significantly different one another, but when the probabilities were compared expressive differences were noticed, indicating the methodology is adequate. It was concluded that the new methodology is a useful alternative for the fitting of diameter distributions and application in modeling of growth and yield of forest stands.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here
Accelerating Research

Address

John Eccles House
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom