Johnson–Schumacher Split-Plot Design Modelling of Rice Yield

Summary In this research, an intrinsically nonlinear split-plot design model (INSPDM) is formulated and studied. It was by fitting a Johnson–Schumacher (JS) function to the mean function. The fitted parameters are estimated using generalized least squares (EGLS) technique based on Gauss–Newton procedure with Taylor series expansion, minimizing objective of model. variance components for whole plot subplot random effects restricted maximum likelihood estimation (REML) techniques. adequacy INSPDM tested four median measures: resistant coefficient determination, prediction modeling efficiency statistic, square error statistic residuals Akaike’s Information Criterion (AIC), Corrected (AICC) Bayesian (BIC) statistics used select best parameter technique. results obtained compared techniques ordinary (OLS) EGLS via (MLE). showed be adequate, reliable, stable, good fit EGLS-REML when OLS EGLS-MLE estimates.