### Abstract

The problem of constructing sequential estimates of the parameters of recursion processes when there are independent noises is solved. It is shown that sequential estimates have the property of uniform boundedness of the mean-square deviation. As an example, the problem of estimating the parameters of an autoregressive process from noisy observations is examined. It is shown that, if the process is stable, then its parameters can be estimated with a given accuracy.

Original language | English |
---|---|

Pages (from-to) | 74-83 |

Number of pages | 10 |

Journal | Engineering cybernetics |

Volume | 20 |

Issue number | 6 |

Publication status | Published - Nov 1982 |

### Fingerprint

### ASJC Scopus subject areas

- Engineering(all)

### Cite this

*Engineering cybernetics*,

*20*(6), 74-83.

**SEQUENTIAL ESTIMATION OF THE PARAMETERS OF DYNAMIC SYSTEMS WHEN THERE IS INCOMPLETE OBSERVATION.** / Vasil'yev, V. A.; Konev, V. V.

Research output: Contribution to journal › Article

*Engineering cybernetics*, vol. 20, no. 6, pp. 74-83.

}

TY - JOUR

T1 - SEQUENTIAL ESTIMATION OF THE PARAMETERS OF DYNAMIC SYSTEMS WHEN THERE IS INCOMPLETE OBSERVATION.

AU - Vasil'yev, V. A.

AU - Konev, V. V.

PY - 1982/11

Y1 - 1982/11

N2 - The problem of constructing sequential estimates of the parameters of recursion processes when there are independent noises is solved. It is shown that sequential estimates have the property of uniform boundedness of the mean-square deviation. As an example, the problem of estimating the parameters of an autoregressive process from noisy observations is examined. It is shown that, if the process is stable, then its parameters can be estimated with a given accuracy.

AB - The problem of constructing sequential estimates of the parameters of recursion processes when there are independent noises is solved. It is shown that sequential estimates have the property of uniform boundedness of the mean-square deviation. As an example, the problem of estimating the parameters of an autoregressive process from noisy observations is examined. It is shown that, if the process is stable, then its parameters can be estimated with a given accuracy.

UR - http://www.scopus.com/inward/record.url?scp=0020202506&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0020202506&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0020202506

VL - 20

SP - 74

EP - 83

JO - Engineering cybernetics

JF - Engineering cybernetics

SN - 0013-788X

IS - 6

ER -