MA to ARMA two-stage Monte Carlo simulation

Marc Mignolet, Pol D. Spanos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A prior autoregressive (AR) model can be the basis to obtain a quite eficient and reliable subsequent autoregressive moving average (ARMA) approximation. The design of the AR system which is straightforward for most processes, can become a quite delicate task when the target spectral matrix belongs to a special class of functions. For this reason an alternative approach to the generation of the required time histories, based on the design of purely moving average (MA) systems has often been used. The goal of the present paper is to combine the apparent robustness of the MA approximation, and the efficiency of the ARMA one, by presenting two two-stage MA to ARMA system design techniques.

Original languageEnglish (US)
Title of host publicationProbab Methods Civ Eng Proc 5th ASCE Spec Conf
Place of PublicationNew York, NY, United States
PublisherPubl by ASCE
Pages265-268
Number of pages4
ISBN (Print)0872626598
StatePublished - 1988
EventProbabilistic Methods in Civil Engineering, Proceedings of the 5th ASCE Specialty Conference - Blacksburg, VA, USA
Duration: May 25 1988May 27 1988

Other

OtherProbabilistic Methods in Civil Engineering, Proceedings of the 5th ASCE Specialty Conference
CityBlacksburg, VA, USA
Period5/25/885/27/88

Fingerprint

Systems analysis
Monte Carlo simulation

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Mignolet, M., & Spanos, P. D. (1988). MA to ARMA two-stage Monte Carlo simulation. In Probab Methods Civ Eng Proc 5th ASCE Spec Conf (pp. 265-268). New York, NY, United States: Publ by ASCE.

MA to ARMA two-stage Monte Carlo simulation. / Mignolet, Marc; Spanos, Pol D.

Probab Methods Civ Eng Proc 5th ASCE Spec Conf. New York, NY, United States : Publ by ASCE, 1988. p. 265-268.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Mignolet, M & Spanos, PD 1988, MA to ARMA two-stage Monte Carlo simulation. in Probab Methods Civ Eng Proc 5th ASCE Spec Conf. Publ by ASCE, New York, NY, United States, pp. 265-268, Probabilistic Methods in Civil Engineering, Proceedings of the 5th ASCE Specialty Conference, Blacksburg, VA, USA, 5/25/88.
Mignolet M, Spanos PD. MA to ARMA two-stage Monte Carlo simulation. In Probab Methods Civ Eng Proc 5th ASCE Spec Conf. New York, NY, United States: Publ by ASCE. 1988. p. 265-268
Mignolet, Marc ; Spanos, Pol D. / MA to ARMA two-stage Monte Carlo simulation. Probab Methods Civ Eng Proc 5th ASCE Spec Conf. New York, NY, United States : Publ by ASCE, 1988. pp. 265-268
@inproceedings{9541477a98884eec9ca6597c9452c032,
title = "MA to ARMA two-stage Monte Carlo simulation",
abstract = "A prior autoregressive (AR) model can be the basis to obtain a quite eficient and reliable subsequent autoregressive moving average (ARMA) approximation. The design of the AR system which is straightforward for most processes, can become a quite delicate task when the target spectral matrix belongs to a special class of functions. For this reason an alternative approach to the generation of the required time histories, based on the design of purely moving average (MA) systems has often been used. The goal of the present paper is to combine the apparent robustness of the MA approximation, and the efficiency of the ARMA one, by presenting two two-stage MA to ARMA system design techniques.",
author = "Marc Mignolet and Spanos, {Pol D.}",
year = "1988",
language = "English (US)",
isbn = "0872626598",
pages = "265--268",
booktitle = "Probab Methods Civ Eng Proc 5th ASCE Spec Conf",
publisher = "Publ by ASCE",

}

TY - GEN

T1 - MA to ARMA two-stage Monte Carlo simulation

AU - Mignolet, Marc

AU - Spanos, Pol D.

PY - 1988

Y1 - 1988

N2 - A prior autoregressive (AR) model can be the basis to obtain a quite eficient and reliable subsequent autoregressive moving average (ARMA) approximation. The design of the AR system which is straightforward for most processes, can become a quite delicate task when the target spectral matrix belongs to a special class of functions. For this reason an alternative approach to the generation of the required time histories, based on the design of purely moving average (MA) systems has often been used. The goal of the present paper is to combine the apparent robustness of the MA approximation, and the efficiency of the ARMA one, by presenting two two-stage MA to ARMA system design techniques.

AB - A prior autoregressive (AR) model can be the basis to obtain a quite eficient and reliable subsequent autoregressive moving average (ARMA) approximation. The design of the AR system which is straightforward for most processes, can become a quite delicate task when the target spectral matrix belongs to a special class of functions. For this reason an alternative approach to the generation of the required time histories, based on the design of purely moving average (MA) systems has often been used. The goal of the present paper is to combine the apparent robustness of the MA approximation, and the efficiency of the ARMA one, by presenting two two-stage MA to ARMA system design techniques.

UR - http://www.scopus.com/inward/record.url?scp=0024130894&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024130894&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0024130894

SN - 0872626598

SP - 265

EP - 268

BT - Probab Methods Civ Eng Proc 5th ASCE Spec Conf

PB - Publ by ASCE

CY - New York, NY, United States

ER -