Osborne, Michael2015-12-081068-9613http://hdl.handle.net/1885/36072A regression problem is separable if the model can be represented as a linear combination of functions which have a nonlinear parametric dependence. The Gauss-Newton algorithm is a method for minimizing the residual sum of squares in such problems. It isKeywords: Algorithms; Boolean functions; Convergence of numerical methods; Curve fitting; Errors; Maximum likelihood estimation; Measurement errors; Mobile telecommunication systems; Newton-Raphson method; Random errors; Consistency; Expected Hessian; Kaufman's mod Consistency; Expected Hessian; Kaufman's modification; Large data sets; Law of large numbers; Maximum likelihood; Newton's method; Nonlinear least; Random errors; Rate of convergence; Scoring; Squares20072016-02-24