Regressão Simplex aplicada a delineamentos de mistura e utilização do Algoritmo Boosting

In the composition of this work are present two parts. The first part contains the theory used. The second part contains the two articles. The first article examines two models of the class of generalized linear models for analyzing a mixture experiment, which studied the effect of different diets consist of fat, carbohydrate, and fiber on tumor expression in mammary glands of female rats, given by the ratio mice that had tumor expression in a particular diet. Mixture experiments are characterized by having the effect of collinearity and smaller sample size. In this sense, assuming normality for the answer to be maximized or minimized may be inadequate. Given this fact, the main characteristics of logistic regression and simplex models are addressed. The models were compared by the criteria of selection of models AIC, BIC and ICOMP, simulated envelope charts for residuals of adjusted models, odds ratios graphics and their respective confidence intervals for each mixture component. It was concluded that first article that the simplex regression model showed better quality of fit and narrowest confidence intervals for odds ratio. The second article presents the model Boosted Simplex Regression, the boosting version of the simplex regression model, as an alternative to increase the precision of confidence intervals for the odds ratio for each mixture component. For this, we used the Monte Carlo method for the construction of confidence intervals. Moreover, it is presented in an innovative way the envelope simulated chart for residuals of the adjusted model via boosting algorithm. It was concluded that the Boosted Simplex Regression model was adjusted successfully and confidence intervals for the odds ratio were accurate and lightly more precise than the its maximum likelihood version.

Rotulagem como mecanismo de compartilhamento de informações com agentes da cadeia produtiva do café

The Agroindustrial System coffee after decades of state regulation has become a decentralized economic activity. This situation compelled most of the actors in the production chain to resort to self r egulatory systems offered by certification agencies, which have specific criteria and procedures. However the diversity criteria of the agencies do not offer resources able to ensure efficient management of the governance structure. There is a need to reduce the asymmetry of information, which is possible through the sharing of data compatible with the production tracking feature. This work is supported in this contextualized problem, based on Transaction Cost Theory and the use of qualitative and quantitative research methodologies. The first part of the work seeks to know the procedures required by leading regulatory agencies in Brazil, with perspective to emphasize the similarities and compatibilities between them, which was reached after analysis of standards and comparison of similarities. Next is registered the operation and the coffee traceability of the structure, in order to identify, through observation, documental and bibliographical analysis, the systems in place and traceable information in line with the recommendations of the coffee certification. Finally, he sought to create a labeling model that allows to generate and share information about the product from the field to the sale for consumption. The information obtained in the previous steps were the basis for data collection with the production chain agents through a structured questionnaire administered to 618 agents of the production chain, whose answers were treated by multivariate methods of statistical analysis. The results showed that actors in the chain are in favor of inclusion and sharing information related to rural production and economic and environmental data. This knowledge enabled develop a labeling model where the information sharing agents is an efficient mechanism to mitigate the governance problems identified.