作者: Murray Aitkin
DOI: 10.1111/J.0006-341X.1999.00117.X
关键词: Expectation–maximization algorithm 、 Mathematics 、 Parametric statistics 、 Statistics 、 Generalized linear model 、 Quasi-likelihood 、 Joint probability distribution 、 Nonparametric statistics 、 Variance function 、 Overdispersion
摘要: … The EM algorithm can be slow to converge in mixture models and does not provide the correct (asymptotic) … Although the Gauss-Newton algorithm might give more efficient model fitting (…