Foi realizada a classificação de coeficientes de variação dos coeficientes de digestibilidade da matéria seca, matéria orgânica, proteína bruta, fibras em detergente neutro e ácido, energia bruta, fibra bruta, celulose, hemicelulose, lignina, extrato etéreo e ganho de peso diário, com base em artigos científicos e teses em eqüinocultura. Os coeficientes de variação foram classificados em: baixo [<(m-1DP)]; médio [(m-1D)<CV<(m+1DP)]; alto [(m+1DP)<CV<(m+2DP)]; e muito alto [>(m+2DP)]. Todas as respostas medidas apresentaram distribuição normal pelo teste Lilliefors. Quanto aos coeficientes de digestibilidade da fibra em detergente neutro e celulose, houve concordância entre a classificação proposta e a sugerida na literatura, sendo os mais discrepantes os coeficientes de digestibilidade da energia bruta e lignina. Desse modo, para cada variável analisada existe distribuição diferenciada dos coeficientes de variação, o que sugere a necessidade de classificação específica para as respostas medidas em nutrição de eqüinos.

Coefficients of variation (CV) distribution for several variables in equine nutrition were obtained. The data are from theses and published scientific papers in equine nutrition. The CV were classified as low [CV£(m-1sd)]; medium [(m-1sd)<CV<(m+1sd)]; high [(m+1sd)<CV<(m+2sd)] and very high [CV>(m+2sd)]. The Lilliefors test for all variables showed no discrepancy from normal distribution. The distributions obtained for coefficients of digestibility for cellulose and fibre in neutral and acid detergent, were close enough to the classification suggested in the literature. However, the distributions for coefficient of digestibility for gross energy and lignin, showed high discrepancies from this classification suggesting that trial precision should be evaluated according to the variable considered. There is a different distribution for coefficients of variation for each variable suggesting the need of specific classification for each characteristic in equine studies.

We develop a procedure for calculating an optimized Discrete Variable Representation (DVR) optimized for a given potential. The method leads to an efficient and compact way to obtain numerical solutions of quantum mechanical problems. The procedure is applied to several physical problems. To illustrate the strength of the algorithm in dealing with multidimensional calculations, we obtain accurate levels up to 19,000 cm-1 for the vibrational energies of the water molecule.