Title : Quality improvement in agronomy by no linear design of experiments and repeated measurements analysis
Abstract:
Mexico has 16 thousand hectares dedicated to the cultivation of ornamental flowers. They produced approximately 83 thousand tons by year. Given the importance of this crop, there is interest on improving the quality of roses by identifying the best cultivars and periods of production. A feature highly valued in this industry is the stem length of the roses (50-70 cm). In this research, generalized Linear models (GLM), generalized linear mixed models (GLMM), generalized additive models (GAM), generalized additive mixed models (GAMM) and vector generalized lineal models (VGLM), were fitted to explore the possible effect of heat units, relative humidity and light over the stem length of two cultivar of roses during two periods. GAM model showed no linearity effect of heat units and relative humidity on stem length. The best conditions to produce plants between 50 and 70 cm were 650 to 830 of heat units and 82.5 to 85 % of relative humidity. GAMM model with repeated measurements ( 1, 7, 14, 21 and 28 days) identified that the best conditions to produced roses with the required characteristics, they were period 1 with cultivar 2. It was concluded that recently developed new statistical models can be very useful to show nonlinear effects "overshadowed" by the indiscriminate use of linear models.
