• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.11.1979.tde-20231122-100346
Document
Author
Full name
Adroaldo Guimarães Rossetti
Institute/School/College
Knowledge Area
Date of Defense
Published
Piracicaba, 1979
Supervisor
Title in Portuguese
Determinação do tamanho ótimo de parcelas em ensaios agrícolas
Keywords in Portuguese
DELINEAMENTO EXPERIMENTAL
EXPERIMENTAÇÃO AGRÍCOLA
MODELOS MATEMÁTICOS
TAMANHO DA PARCELA
Abstract in Portuguese
O estudo do tamanho ótimo de parcelas, na experimentação de campo, visa a minimizar os custos, em função do erro experimental. As inúmeras pesquisas existentes na literatura indicam sua grande importância. A principal questão envolvida é a heterogeneidade do solo que influi, em grande escala, no estabelecimento do tamanho ideal da parcela. Neste trabalho estima-se o coeficiente de regressão b , ou índice de variabilidade do solo, proposto por FAIRFIELD SMITH (1938), partindo-se de um modelo de regressão em que os dados são correlacionados e têm variâncias desiguais, minimizando-se a soma dos quadrados dos erros através do método dos quadrados mínimos generalizados. Tal estimativa, (Descrito na Dissertação), bem como as estimativas das variâncias observadas, dos diferentes tamanhos de parcelas, são ponderadas pelos pesos wij. Desenvolve-se uma metodologia que visa a estimar eficientemente os pesos wij dos próprios dados. Tal procedimento é válido tanto para dados de ensaios de uniformidade, aqui estudados através de um esquema de classificação hierárquica, ou seja, yijkl = m + ri + bij + pijk+sijkl como para dados de experimentos, tipo comparativo de variedades, onde haja efeitos de tratamentos, conduzido aqui através da análise de um experimento em delineamento de blocos ao acaso com parcelas subdivididas, ou seja yijkl = m + ri + tj + εlij + tk’ + (t t’)jk + ε2ijk + eijkl Demonstra-se que a estimativa b̂, obtida por tal procedimento tem variância mínima assintótica. O desvio de regressão, (Descrito na Dissertação)
Title in English
Not available
Abstract in English
The study of the optimum plot size in field experiments aims at minimizing costs, in terms of experimental error. The great number of researches existing in the literature indicate its great importance. The principal issue involved is the soil heterogeneity which exerts an influence in the determination of the optimum plot size. ln this study, the regression coefficient b or soil variability index proposed by FAIRFIELD SMITH (1938) was estimated using a regression model in which the data are correlated and have unequal variances, minimizing the sum of squares of the errors through the general ized least square method. Such estimate, (See Dissertation) as well as the variance estimates observed, of the different plot sizes, are weighted by the wij weights. The methodology proposed by HATHEWAY and WILLIAMS (1958) is used, with the aim of effectively estimating the wij weight of the data. Such procedure is used for uniformity trial data, studied, in this case, through a hierarchic classification model, that is yijkl = m + ri + bij + pijk+sijkl as well as for experimental data, of the comparative variety type, where there are treatment effects in this case conducted through the analysis of an experiment in randomized block design, with split-split plots, that is yijkl = m + ri + tj + εlij + tk’ + (t t’)jk + ε2ijk + eijkl lt is proved that the estimate of b, obtained through this procedure has an asymptotic minirnurn variance. The regression deviation, (See Dissertation)
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2023-11-24
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.