No label defined (Q3685903)
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 3909587
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 3909587 |
Statements
1984
0 references
finite-length robust predictors
0 references
time series
0 references
time domain approach
0 references
ARMA models
0 references
stationary stochastic processes
0 references
filtering
0 references
prediction
0 references
partial knowledge
0 references
minimax criterion
0 references
least favorable case under uncertainty
0 references
nonlinear estimation
0 references
minimum mean-square error
0 references
minimax-robust
0 references
Methods for constructing minimax filters
0 references
0.8437100052833557
0 references
0.8141739964485168
0 references
0.8119052648544312
0 references
0.8041089177131653
0 references
[For the entire collection see Zbl 0553.00008.] This paper is a synthesis of some previous results, most of them due to the authors, complementing, enhancing, discussing and exemplifying them. It deals with aspects of linear and nonlinear filtering and prediction under partial knowledge about the process. The robust approach is based on the minimax criterion and on the suitably defined least favorable case under uncertainty. Methods for determining the least favorable cases are given and discussed. The results are derived for both cases of spectral and probabilistic uncertainty. In the last case, general results are given on nonlinear estimation, based on the minimum mean-square error. It is proved that linear filters are minimax-robust under all filters if only information about the second moments are available. Again, a deep discussion of different cases is given. Methods for constructing minimax filters are analysed under some particular conditions. The emphasis on both generality and applicability is one of the merits of this paper. For those interested in applications, the paper reviewed below, Zbl 0569.62084, is a valuable reading.
0 references