Answers are grounded in this paper and cite their sources.
We propose that given a data-set S=\(xᵢ,yᵢ)/i=1,2,n\ and real-valued functions _α(x)/α=1,2,m\, the least-squares fit vector A=_α\ for y=∑_α aαf_α(x) is A = (FTF)⁻¹FTY where [Fiα]=[f_α(xᵢ)]. We test this formalism by deriving the algebraic expressions of the regression coefficients in $y = ax + b$ and in y = ax² + bx + c. As a practical application, we successfully arrive at the coefficients in the semi-empirical mass formula of nuclear physics. The formalism is { generic} - it has the potential of being applicable to any { type} of ᵢ\ as long as there exist appropriate _α\. The method can be exploited with a CAS or an object-oriented language and is excellently suitable for parallel-processing.
No takes yet. Share an insight, caveat, or question.
Upadhyay et al. (2013) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: