作者: P. J. Bramel , P. N. Hinz , D. E. Green , R. M. Shibles
DOI: 10.1007/BF00021136
关键词: Function (mathematics) 、 Statistics 、 Covariance matrix 、 Botany 、 Variables 、 Correlation 、 Biology 、 Raceme 、 Linear regression 、 Indeterminate 、 Set (abstract data type)
摘要: The study was conducted to identify plant characters associated with seed yield in close soybean spacings. Lines selected from two F6 populations the F10 generation segragating for degree of stem termination were grown locations. Traits measured included lengths developmental stage. and canopy height, number nodes, lodging at different stages, raceme lengths, nodes branch nodes. One problem analyzing data drawing conclusions such a is related complex nature interrelationships large traits. Another, not unrelated problem, involves calculation multiple-regression equations multicolinearities. Because these problems, factor analysis used correlation matrix sets variables same biological concept or function. within determined each type, measurements standardized. Means calculated by using sum standardized set. Multiple regression relationships between as independent dependent variable. Individual traits set substituted multiple mean value With use this technique determine equations, resulting involved functions instead repeated function plant. It found that, determinate measurement seed-filling period could possibly be predict while, semideterminate ‘fixed capital’ terminal length useful. Finally, indeterminate used, but it had very low predictive value.