|
|
||||||||
Utah State University, Logan, Utah 84321
Abstract
Concepts of Regression Equations
WOOD and Capstick (1926), Titus (1928), Brody and Proctor (1935), and Blaxter and Wood (1951) were among early nutritionists to use regression equations to estimate nutrient requirements. Schneider et al. (I952) devised regression equations to predict the digestibility of feeds. The method of using regression equations, however, has not been widely used as a means of solving practical nutrition problems.
A significant regression or change of one variable in relation to another indicates a strong possibility of predicting an unknown value from one that is known.
By the usual statistical convention the dependent variable is designated as Y (Ostle, 1954), and the independent variables are designated as X. Each independent variable (X) included in the equation will fluctuate with Y in its own unique way.
Linear regression equations may be written in the form: Y=b0+b1X1+b2X2+b3X3. . . etc. where b0 is the Y intercept when X=O (regression constant) and bl, b2, b3, . . . etc. are the respective changes in Y per unit change (regression coefficient) of independent variables Xl, X2, X3 . . . . etc.
1 Report on project 604, Utah Agricultural Experiment Station, Journal Series No. 1235.
2 Presented at the 63rd Annual Meeting of the American Society of Animal Science, University of California, Davis, August 4, 1971, as part of a Symposium on Biological Availability of Nutrients in Feeds. Co-sponsored by the A.S.A.S. and the Committee on Animal Nutrition, N.A.S./N.R.C.
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |