About Gaussian schemes in stochastic approximation (Q1318334)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 540197
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | About Gaussian schemes in stochastic approximation |
scientific article; zbMATH DE number 540197 |
Statements
About Gaussian schemes in stochastic approximation (English)
0 references
27 April 1994
0 references
The author considers a continuous-time analogue of the linear version of the Robbins Monro algorithm of stochastic approximation, where the error process is a Gaussian martingale. Regarding the asymptotic behavior of the estimates and costs of the procedure, almost-sure and weak convergence results are obtained. The derivations depend on limit theorems on Gaussian martingales and their associated quadratic functionals and an asymptotic property of functions of finite variation which are of independent interest.
0 references
rates of convergence
0 references
strong consistency
0 references
almost sure convergence
0 references
Robbins Monro algorithm
0 references
error process
0 references
costs
0 references
weak convergence
0 references
limit theorems on Gaussian martingales
0 references
quadratic functionals
0 references
asymptotic properties of functions of finite variation
0 references
0 references
0 references
0.8565827012062073
0 references
0.8336816430091858
0 references
0.8118846416473389
0 references
