Optimum plot for coffee culture obtained by data simulation with known variances

Abstract

Several methods have been used by researchers to determine the optimum size of experimental plot and control of experimental error, the most widespread are: empirical method of Smith (1938), method of visual inspection of maximum curvature (Federer, 1955), method of maximum curvature coefficient of variation (Lessman & Atkins, 1963). Some methods of segmented model adjustment was recently proposed to (linear with plateau) to determine the optimum plot size (Paranaiba et al., 2009). All methods mentioned above have used data from assays for measuring uniformity variability between possible plot sizes. The data used in this study were obtained through computer simulation data of a uniformity test. The goal proposed in this paper is to use statistical techniques to determine the optimal sample size. We conclude the optimum plot size was achieved by the method of maximum curvature coefficient of variation which showed a R2 of 0.9818

Modelos de regresión segmentada para estimar el tamaño óptimo de la parcela experimental con caña de azúcar

“…Peixoto et al (2011) en ensayos con maracuyá le atribuyen esta tenencia a la curvatura del modelo. Mendes et al (2016) encontraron un tamaño de parcela para café de 10.53 UEB más grande con el método QRP respecto al método LRP. Por otro lado, Silva et al (2012) utilizando un ensayo de rábano, encontraron una diferencia de 2.57 UEB a favor de método QRP.…”

“…Otra alternativa dentro de las posibilidades de los modelos segmentados es el método de regresión cuadrática con constante (QRP), descrito por Ferreira (2007) y aplicado por Mendes et al (2016) para la estimación del tamaño de la parcela experimental. Con respecto al modelo anterior, este método asume una forma polinomial de segundo grado, en lugar de una forma lineal en el primer segmento (Figura 2).…”

“…Investigaciones actuales fueron en gran parte desarrollados en Brasil y aplicados en diversos cultivos, por ejemplo: crotalaria (Crotalaria juncea L.) (Facco et al, 2017), berenjena (Solanum melongena) (Krysczun et al, 2018), tomate cherry (Solanum lycopersicum L.) (Giacomini y Lúcio, 2018), tomate (Solanum lycopersicum), habichuelas (Phaseolus vulgaris) y calabacín (Cucurbita pepo) (Schwertner et al, 2015), girasol (Helianthus annus L.) (Santos et al, 2015), café (Coffea) (Mendes et al, 2016), taro (Colocasia esculenta) (Silva, 2014) y boniato (Ipomoea batatas) (Rodríguez et al, 2018).…”

Tamanho ótimo de parcela em aveia branca com comparação de três métodos

“…Studies have shown decreasing estimates of Xo in the following order: QRP, LRP and MMC (Silva et al, 2012;Moreira et al, 2016;González et al, 2018;Guimarães et al, 2019;Cargnelutti et al, 2021a,b); higher estimates of Xo by QRP compared to LRP (Peixoto et al, 2011); and higher estimates of Xo by LRP compared to MMC (Brito et al, 2012;Leonardo et al, 2014;Guarçoni et al, 2017;Sousa et al, 2018;Brioschi et al, 2020). Therefore, in these studies with the approach of comparing methods to determine the optimal plot size, results similar to those of the present study were found.…”

“…The values of CV (X) as a function of X can be related by the methods of modified maximum curvature (MMC) (Meier & Lessman, 1971), linear response and plateau model (LRP) (Paranaíba et al, 2009) and quadratic response and plateau model (QRP) (Peixoto et al, 2011), and make it possible to determine the optimal plot size (Xo) and the coefficient of variation in the optimal plot size (CV Xo ). Comparative studies involving the MMC and LRP methods have been carried out with papaya (Brito et al, 2012), pineapple (Leonardo et al, 2014), cabbage (Guarçoni et al, 2017) and cassava (Sousa et al, 2018), the LRP e QRP methods with passion fruit (Peixoto et al, 2011) and MMC, LRP and QRP methods with radish (Silva et al, 2012), sweet potato (González et al, 2018), cactus pear (Guimarães et al, 2019), coffee (Moreira et al, 2016, Brioschi et al, 2020, millet + slender leaf rattlebox + showy rattlebox (Cargnelutti et al, 2021a) and buckwheat (Cargnelutti et al, 2021b), evidencing distinct results between the methods and the importance of using more than one method to determine the optimal plot size.…”