Abstract  An informal discussion is given on performing an unconstrained maximization or solving non‐linear equations of statistics by iterative methods with the quadratic termination property. It is shown that if a miximized function, e.g. likelihood, is asymptotically quadratic, then for asymptotically efficient inference finitely many iterations are needed. Copyright

asymptotically differentiable, asymptotically quadratic, conjugate gradient, Davidon‐Fletcher‐Powell, methods of Newton‐Raphson, quadratic termination, quasi‐Newton, scoring
dx.doi.org/10.1111/j.1467-9574.1983.tb00813.x
Statistica Neerlandica

Dzhaparidze, K.O. (1983). On iterative procedures of asymptotic inference. Statistica Neerlandica, 37(4), 181–189. doi:10.1111/j.1467-9574.1983.tb00813.x