Problems of Convergence of Maximum Likelihood Iterative Procedures in Multiparameter Situations

摘要: Abstract In multiparameter situations there are important problems of convergence in the solution maximum likelihood iterative equations. Convergence may be critically dependent on initial parameter estimates employed. These indicated specifically for a model exponentially-distributed lifetimes where reciprocal exponential is linearly various regressor variables. A strategy suggested based alternative use observed or expected values sample second derivative log likelihood. For examples shown correspond to no effect variables, this being equivalent some cases making an unweighted fit data.