
handle: 11449/34504
Em trabalho recente sobre ajuste de peso de vacas leiteiras em função do tempo, Kroll (1990) ajustou várias funções, entre elas a de Gompertz, com erros auto-regressivos. No entanto, o ajuste sem considerar a autocorrelação explicou a mais até 25% da variação total do que o modelo de erros autocorrelacionados. Neste trabalho, o modelo de Gompertz com erros auto-regressivos é retomado com acréscimo do efeito de estação do ano. A sugestão dessa inclusão foi baseada na função de autocorrelação da série original, depois de descontada a tendência. Para estimação dos seis parâmetros, no processo iterativo foi utilizada uma matriz = [ | ], onde é a mesma matriz do método de Stevens para ajuste de Gompertz, e , a matriz de 0s, 1s e -1s, para sazonalidade. Como resultado, houve elevação do coeficiente de determinação em relação ao maior valor obtido no trabalho acima citado.
In recent work (Kroll, 1990) on dairy cattle growth using its monthly weights, a Gompertz function, first considering the autocorrelated error structure, and second, the independent error structure, was fitted. The last explained 25% more of the total variance than the former. In this work, a Gompertz function was fitted with autocorrelated error structure adding seasonality to the model. This was suggested taking into account the time series autocorrelation function. In order to estimate the six parameters, an iterative procedure was used, employing a matrix X = [G\S], where G is the usual matrix for fitting the Gompertz function, and S is a 0s, 1s, -1s matrix necessary to take into account seasonality. With this model, the adjustment was better in terms of the coefficient determination.
NONLINEAR REGRESSION, ERROR AUTOCORRELATION, GOMPERTZ FUNCTION
NONLINEAR REGRESSION, ERROR AUTOCORRELATION, GOMPERTZ FUNCTION
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 0 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
