Premium
Relationship Between Soybean Yields and Leaf Levels of 10 Elements Determined with Different Regression Models 1
Author(s) -
Walker W. M.,
Peck T. R.,
Carmer S. G.
Publication year - 1969
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj1969.00021962006100030024x
Subject(s) - polynomial regression , mathematics , regression analysis , linear regression , regression , polynomial , stepwise regression , statistics , quadratic equation , mathematical analysis , geometry
Soybean leaf samples and yields were obtained from experimental plots receiving varying levels of P and K. Yields were regressed on a quadratic polynomial with leaf levels of P and K as variables. This regression had an R 2 of 0.495. Yields were regressed on a second quadratic polynomial with leaf levels of N, P, K, Ca, Mg, B, Cu, Fe, Mn, and Zn as variables. With the stepwise regression procedure used for fitting the model, 22 linear and second order terms were fitted to the data. This regression had an R2 of 0.813. The additional variation accounted for by the second regression was statistically significant. Other polynomial regressions were fitted to the data using ratios of elements as independent variables, but none were superior to the quadratic polynomial in terms of R 2 values.