Comparison of corn yield regression models derived from stratified and statewide soil, weather, and management data in Iowa

Abstract

Data sets developed in the factorial stratification by AREA and TIME 32 Symbols and identification of the variables included in the corn yield regressions on soil, weather, and management variables 35 Means and ranges of the variables included in the yield regressions on soil, weather, and management variables, stratified data for ETl, ET2, WTl, and WT2 series of models 60 Simple correlation coefficients between yield, management, climatic, and soil variables in the ETl, ET2, WTl, and WT2 data sets 63 Variates included in the regressions of corn yield on the linear or quadratic functions of selected variables. Models QETl and QET2 series 72 Variates included in the regressions of com yield on the linear or quadratic functions of selected variables. Models QWTl and QWT2 series 73 Summary of the model selection steps. Models QETl, QET2, QWTl, and QWT2 series 74 Regression statistics of the final quadratic models of com yield on selected variates. Models QETl-25 and QET2-17 77 Regression statistics of the rinal quadratic models of corn yield on selected variates. Models QWTl-17 and QWT2-15 79 Relative changes in yield (AY) for variables with significant quadratic terms, Models QETl-25, QET2-17, QWTl-17, and QWT2-15a 82 Means and ranges of the variables included in the yield regressions on soil, weather, and management variables, stratified data for EAST, WEST, TIMEl, and TIME2 series of models 100

Simulation of kernel development and assimilate partitioning in maize

“…1. regression analyses may not reflect true cause/effect relationships among the variables studied 2. the models may be subjective (based on the variables the modeler chooses to include) 3. predictions based on regression analyses are valid only over the range of conditions (temporal, geographic and cultural) in which the original data were collected (Macias Laylle, 1984;Olson and Olson, 1986;Ortiz Valdéz, 1985).…”