Generalized Coefficient of Determination
暂无分享,去创建一个
Abstract In the theory of cost-estimating relationship (CER) development using the method of ordinary least-squares (OLS) linear regression, where the dependent variable is y (e.g., cost) and the independent variable is x (e.g., weight, power, thrust, etc.), the square of the correlation coefficient is called the “coefficient of (linear) determination.” Usually denoted by the symbol R 2, the coefficient represents the proportion of variation in y that can be explained by passing variations in x up through the linear relationship. As such, it is often interpreted as providing a measure of the quality of the CER as a predictor of cost. The present report offers a generalized definition of the coefficient of determination that is valid for comparing the quality of non-linear versus linear CERs and has none of the well known drawbacks, e.g., those related to the logarithmic transformation, of other proposed definitions.
[1] Transformations and R 2 , 1991 .
[2] L. Magee,et al. R 2 Measures Based on Wald and Likelihood Ratio Joint Significance Tests , 1990 .
[3] T. O. Kvålseth. Cautionary Note about R 2 , 1985 .
[4] Richard F. Gunst,et al. Applied Regression Analysis , 1999, Technometrics.
[5] Philip H. Young. The Meaning of the Standard Error of the Estimate in Logarithmic Space , 1999 .