ABSTRACT
Objective To compare the Bayesian statistics and the classical statistics in the quantile regression analysis in order to select a more effective method .Methods The large sample data was chosen , and the QUANTREG procedure in SAS was used for the classical statistics and the MCMC procedure in SAS for the Bayesian one , respectively .Using ten-fold cross-validation method , the goodness of fitting of the models was appraised in terms of the fitted effect based on the training dataset and the predicted effect based on the predictive dataset .Results In most cases, the indexes of the quantile regression models in the classical statistics were slightly worse than those of the Bayesian one .In the ten-fold cross-validation of the partial samples as a training dataset , the fitting effect of the lower quartile ( Q1 ) and upper quartile ( Q3 ) of the Bayesian statistics was better than that of the classical one .However , the median ( Q2 ) fitting effect of the Bayesian statistics was slightly worse than that of the classical one .As for the prediction effect , the Bayesian statistical quantile regression model was superior to the classic one .Conclusion To expect high accuracy , such as the predictive effects and fitting effects of each quantile , the Bayesian quantile regression analysis should be chosen .If the major concern is the fitting effect of the median , careful selection from the approaches mentioned above is needed .If time and energy are limited, and the sample size is large enough , the classic statistical quantile regression analysis is a good choice .
ABSTRACT
Na operação de semeadura realizada com semeadoras-adubadoras inúmeros fatores interferem no estabelecimento do estande de plantas, dentre estes, a distribuição longitudinal das sementes em função da velocidade de deslocamento pode afetar significativamente a produtividade das culturas. O objetivo do presente trabalho foi avaliar a distribuição longitudinal de sementes de milho em duas velocidades de deslocamento de uma semeadora-adubadora de precisão, através da estatística clássica e geoestatística. Para isso, foi construída uma malha de amostragem em grid regular com distância entre os pontos de 10 m, totalizando 100 pontos amostrais em cada área. Cada ponto amostral constituiu uma área de 3,6 m2 onde se mediu a percentagem de espaçamentos aceitáveis, duplos e falhos entre plantas, após semeadura nas velocidades de deslocamento de 4,58 e 5,94 km h-1 . Os dados foram analisados pela estatística clássica e geoestatística. A percentagem de espaçamentos aceitáveis e falhos apresentou diferença significativa entre as velocidades. Os resultados indicaram ausência de dependência espacial para as percentagens de espaçamentos estudados nas duas velocidades, indicando que estudos e inferências estatísticas podem ser realizados com base em parâmetros da estatística clássica para distância maior que a menor utilizada na amostragem.
In the operation of sowing with planter numerous factors interfere with the establishment of the plant stand, among them, the longitudinal distribution of seeds depending on the displacement speed can significantly affect crop productivity. The present study aimed the evaluation of longitudinal distribution of maize seeds in two different speeds of dislocation of a precision grain drill, through classical and geostatistics. A regular grid sample, totalizing 100 points in each area, was built with 10-meter distance among the points. Each sample point comprised 3.6m2 (1.8 x 2.0) where the percentage of acceptable, double and fail spacing among the plants was measured after sowing at 4.58 and 5.94 km h-1 speed of dislocation. Classical and geostatistics were used for data analyses. The percentage of acceptable and failed spacing has shown significant difference between 4.58 and 5.94 km h-1 speed. The results has shown absence of spatial dependence regarding the percentage of the studied spacing (acceptable, fail and double) at both speeds, showing that studies and statistical inferences can occur based on parameters of classical statistics for distances higher than the shortest one used in the sampling.