NotesFAQContact Us
Collection
Advanced
Search Tips
Peer reviewed Peer reviewed
ERIC Number: EJ453857
Record Type: Journal
Publication Date: 1992
Pages: N/A
Abstractor: N/A
Reference Count: N/A
ISBN: N/A
ISSN: ISSN-0033-3123
Computing Maximum Likelihood Estimates of Loglinear Models from Marginal Sums with Special Attention to Loglinear Item Response Theory.
Kelderman, Henk
Psychometrika, v57 n3 p437-50 Sep 1992
Describes algorithms used in the computer program LOGIMO for obtaining maximum likelihood estimates of the parameters in loglinear models. These algorithms are also useful for the analysis of loglinear item-response theory models. Presents modified versions of the iterative proportional fitting and Newton-Raphson algorithms. Simulated data illustrate the algorithms. (SLD)
Publication Type: Journal Articles; Reports - Evaluative
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A