Claudio Guilherme Portela de Carvalho

Análise de trilha sob multicolinearidade em pimentão

The main goals of this work were: a) to display genotype correlations in direct and indirect effects of primary and secondary components on the production of fruits, in green pepper (Capsicum annuum L.) genotypes, by mean of path analysis; b) to compare alternative methods of estimation of the path coefficients to avoid the adverse effects of multicollinearity. Based on the magnitude of direct and indirect effects, none of the primary or secondary components showed to be more important in the selection of more productive genotypes. In segregating populations, the character height of the plant might be useful in correlated response, by means of indirect selection. In the occurrence of moderate to severe multicollinearity between the characters, the elimination of variables from the regression model and the ridge path analysis showed to be efficient in reducing the high variances of the path coefficients found. However, when several variables need to be eliminated, the disposal process becomes difficult and many analysis might be necessary. Ridge path analysis, although, should by-pass the adverse effects of multicollinearity, without having great reduction on the coefficient determination values of the regression models.

Estimating gain by use of a classic selection index under multicollinearity in wheat (Triticum aestivum)

It was shown that the classic selection index, under multicollinearity, could not give simultaneous gains for wheat grain production and its primary components. This was due to the instability and, consequently, low precision of the coefficient index estimates. A modification of the prediction process of the index was proposed to avoid the adverse effects of multicollinearity, adopting a procedure based on ridge regression theory. The modified classic selection index, or ridge index, gave more statistically viable index coefficient estimates and gains for all of the characters evaluated. However, lower gains for number of grains per spike and grain yield were obtained, when compared to those obtained with selection for grain yield.