Least-squares fit of a linear combination of functions

We propose that given a data-set and real-valued functions the least-squares fit vector for is where We test this formalism by deriving the algebraic expressions of the regression coefficients in and in 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 as long as there exist appropriate The method can be exploited with a CAS or an object-oriented language and is excellently suitable for parallel-processing.