|
Associate Prof. Evrim ORAL |
|
Louisiana State University Health Sciences Center School of Public Health, Biostatistics Program |
||
The method of modified maximum likelihood (MML) estimation was originated by M. L. Tiku (1967, 1978, 1980) and has been used extensively in literature (Tan and Tabatabai, 1988; Tiku and Suresh 1992; Oral and Gunay, 2004, Oral 2006; Oral and Oral 2011; Oral 2012). The methodology of MML is employed in situations where the maximum likelihood (ML) estimation is intractable. There are three steps to apply the method: (i) express the likelihood equations in terms of the order statistics, (ii) replace the intractable functions by their linear approximations such that the differences between the two converge to zero as n tends to infinity, and (iii) solve the resulting equations. The solutions, called MML estimators (MMLEs), have closed forms, and are therefore easy to compute. A proof is available in Vaughan and Tiku (2000) for the fact that, under some regularity conditions, MMLEs have exactly the same asymptotic properties as ML estimators (MLEs), and for small n values they are known to be essentially as efficient as MLEs.
|
Contact Information: |
|
LSUHSC School of Public Health 2020 Gravier Street 3rd Floor New Orleans, LA 70112 |
||
Phone: (504) 568-6094 Fax: (504) 568-5701 e-mail: eoral@lsuhsc.edu |