What can or can't be estimated in branching and related processes? (Q1263206)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: What can or can't be estimated in branching and related processes? |
scientific article; zbMATH DE number 4126528
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | What can or can't be estimated in branching and related processes? |
scientific article; zbMATH DE number 4126528 |
Statements
What can or can't be estimated in branching and related processes? (English)
0 references
1989
0 references
This note compares the incompletely observed random walk and Galton- Watson branching tree. For the random walk, parameters not completely determined by the first few moments of the underlying distribution cannot be consistently estimated. A similar result holds for the branching tree with the observations now being sums of family sizes. In the case where the offspring distribution belongs to a power series family, it is possible to obtain the maximum likelihood estimators from an approximate likelihood. The estimates are the same as when individual family sizes are observed.
0 references
consistent estimation
0 references
incompletely observed random walk
0 references
Galton-Watson branching tree
0 references
sums of family sizes
0 references
offspring distribution
0 references
power series family
0 references
maximum likelihood estimators
0 references
approximate likelihood
0 references
0.83013254
0 references
0.82992554
0 references
0.82764745
0 references
0.80598474
0 references