ABSTRACT
ABSTRACT: The objective of this paper was to identify the direct and indirect effects of feijoa fruits (Acca sellowiana) traitson pulp weight, in order to use these traits in indirect genotypes selection. Fruits of five feijoa plants were collected in Rio Grande do Sul, in the years of 2009, 2010 and 2011. Six traits were evaluated: diameter, length, total weight, pulp weight, peel thickness and number of seeds per fruit. In the path analysis, with or without ridge regression, pulp weight was considered as the basic variable, and the other traits were considered as explanatory variables. Total weight and fruit diameter had high direct effect, and are the main traits associated with pulp weight. These traits may serve as criteria for indirect selection to increase feijoa pulp weight, since they are easy to be measured.
RESUMO: O objetivo deste trabalho foi identificar os efeitos diretos e indiretos de caracteres dos frutos de goiabeira-serrana (Accasellowiana) sobre o peso de polpa, visando à utilização desses caracteres para a seleção indireta de genótipos. Foram coletados frutos de cinco grupo de plantas de goiabeira-serrana no Rio Grande do Sul, nos anos de 2009, 2010 e 2011. Seis caracteres foram avaliados: diâmetro, comprimento, peso total, peso de polpa, espessura da casca e número de sementes por fruto. Na análise de trilha com e sem regressão em crista para controle da multicolinearidade, o peso de polpa foi considerado variável básica e os demais caracteres considerados explicativos. O peso total, seguido do diâmetro de fruto, tiveram efeito direto elevado e são os principais caracteres associados ao peso de polpa. Estes caracteres podem servir de critérios de seleção indireta para aumentar o peso de polpa de goiabeira-serrana, pois são de fácil mensuração.
ABSTRACT
We analysed geographic variation in the number of males per 100 females at birth also called the secondary sex ratio (SR), the crude birth rate (CBR), the total fertility rate (TFR) and the twinning rate (TWR). Earlier studies have noted geographic variations in the TWR and racial differences in the SR. Statistical analyses have shown that comparisons between SRs demand large data sets because random fluctuations in moderate data are marked. Consequently, reliable results presuppose national birth data. In this study, we analysed historical demographic data and investigated the geographic variations between the counties in Sweden for the SR among the live born (1749-1869), the CBR in 1751-1870, the TFR in 1860 and the TWR in 1751-1860. We built spatial models and as regressors we used geographic co-ordinates for the residences of the counties in Sweden. The influence of the CBR and TFR on the SR and TWR was examined. For all variables, we obtained spatial variations, albeit of different patterns and power. Hence, no common spatial pattern for the demographic variables SR, TFR, CBR and TWR was detected, but a better fit was noted for TFR, CBR and TWR than for SR.
ABSTRACT
The problem of multicollinearity in regression analysis was studied. Ridge regression (RR) techniques were used to estimate parameters affecting the performance of crossbred calves raised in tropical and subtropical regions by a model including additive, dominance, joint additive or "profit heterosis" and epistatic effects and their interactions with latitude in an attempt to model genotype by environment interactions. A software was developed in Fortran 77 to perform five variant types of RR: the originally proposed method; the method implemented by SAS; and three methods of weighting the RR parameter lambda. Three mathematical criteria were tested with the aim of choosing a value for the lambda coefficient: the sum and the harmonic mean of the absolute Student t-values and the value of lambda at which all variance inflation factors (VIF) became lower than 300. Prediction surfaces obtained from estimated coefficients were used to compare the five methods and three criteria. It was concluded that RR could be a good alternative to overcome multicollinearity problems. For all the methods tested, acceptable prediction surfaces could be obtained when the VIF criterion was employed. This mathematical criterion is thus recommended as an auxiliary tool for choosing lambda.