Reconsideration of dimensions and curve fitting practice in view of Georgescu-Roegen's epistemology in economics

This paper is to examine the proper use of dimensions and curve fitting practices elaborating on Georgescu-Roegen's economic methodology in relation to the three main concerns of his epistemological orientation. Section 2 introduces two critical issues in relation to dimensions and curve fitting practices in economics in view of Georgescu-Roegen's economic methodology. Section 3 deals with the logarithmic function (ln z) and shows that z must be a dimensionless pure number, otherwise it is nonsensical. Several unfortunate examples of this analytical error are presented including macroeconomic data analysis conducted by a representative figure in this field. Section 4 deals with the standard Cobb-Douglass function. It is shown that the operational meaning cannot be obtained for capital or labor within the Cobb-Douglas function. Section 4 also deals with economists' "curve fitting fetishism". Section 5 concludes this paper with several epistemological issues in relation to dimensions and curve fitting practices in economics.