Existence and uniqueness of the solution of the likelihood equations for binary Markov chains
The two-state Markov chain is a useful model when analyzing binary data that may be serially correlated. The equations likelihood equations for this model are two intersecting conics, and several solutions could be anticipated. However, we prove the existence and uniqueness of the solution, and give a simple method for numerical solution.
Year of publication: |
1991
|
---|---|
Authors: | Bisgaard, Søren ; Travis, Laurel E. |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 12.1991, 1, p. 29-35
|
Publisher: |
Elsevier |
Keywords: | Binary data binary Markov chain maximum likelihood estimation modeling serial dependence |
Saved in:
Saved in favorites
Similar items by person
-
Simulation for decision making
Thesen, Arne, (1992)
-
A generalization of the alias matrix
Kulahci, Murat, (2006)
-
Blocking Generators for Small 2k-p Designs
Bisgaard, Søren, (1994)
- More ...