Multi-event decision making over multivariate time series

Chun Kit Ngan, Alexander Brodsky, Jessica Lin

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

We propose a multidimentional time-point model and algorithm to solve multi-event expert query parametric estimation (ME-EQPE) problems over multivariate time series. Our proposed model and algorithm combine the strengths of both domain-knowledge-based and formal-learning-based approaches to learn optimal decision parameters for maximising utility over multivariate time series. More specifically, our approach solves the decision optimisation problems to maximise the utility from multiple decision time points, as well as maintaining an optimality of the learned multiple sets of decision parameters in their respective events during the computations. We show that our approach produces a reasonable forecasting result by using the learned multiple sets of decision parameters.

Original languageEnglish (US)
Pages (from-to)263-282
Number of pages20
JournalInternational Journal of Information and Decision Sciences
Volume5
Issue number3
DOIs
StatePublished - Dec 1 2013

Fingerprint

Time series
Decision making
Multivariate time series

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Strategy and Management
  • Information Systems and Management
  • Management of Technology and Innovation

Cite this

Ngan, Chun Kit ; Brodsky, Alexander ; Lin, Jessica. / Multi-event decision making over multivariate time series. In: International Journal of Information and Decision Sciences. 2013 ; Vol. 5, No. 3. pp. 263-282.
@article{c8191cb1832d45de9941f178f8768346,
title = "Multi-event decision making over multivariate time series",
abstract = "We propose a multidimentional time-point model and algorithm to solve multi-event expert query parametric estimation (ME-EQPE) problems over multivariate time series. Our proposed model and algorithm combine the strengths of both domain-knowledge-based and formal-learning-based approaches to learn optimal decision parameters for maximising utility over multivariate time series. More specifically, our approach solves the decision optimisation problems to maximise the utility from multiple decision time points, as well as maintaining an optimality of the learned multiple sets of decision parameters in their respective events during the computations. We show that our approach produces a reasonable forecasting result by using the learned multiple sets of decision parameters.",
author = "Ngan, {Chun Kit} and Alexander Brodsky and Jessica Lin",
year = "2013",
month = "12",
day = "1",
doi = "10.1504/IJIDS.2013.055925",
language = "English (US)",
volume = "5",
pages = "263--282",
journal = "International Journal of Information and Decision Sciences",
issn = "1756-7017",
publisher = "Inderscience Publishers",
number = "3",

}

Multi-event decision making over multivariate time series. / Ngan, Chun Kit; Brodsky, Alexander; Lin, Jessica.

In: International Journal of Information and Decision Sciences, Vol. 5, No. 3, 01.12.2013, p. 263-282.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Multi-event decision making over multivariate time series

AU - Ngan, Chun Kit

AU - Brodsky, Alexander

AU - Lin, Jessica

PY - 2013/12/1

Y1 - 2013/12/1

N2 - We propose a multidimentional time-point model and algorithm to solve multi-event expert query parametric estimation (ME-EQPE) problems over multivariate time series. Our proposed model and algorithm combine the strengths of both domain-knowledge-based and formal-learning-based approaches to learn optimal decision parameters for maximising utility over multivariate time series. More specifically, our approach solves the decision optimisation problems to maximise the utility from multiple decision time points, as well as maintaining an optimality of the learned multiple sets of decision parameters in their respective events during the computations. We show that our approach produces a reasonable forecasting result by using the learned multiple sets of decision parameters.

AB - We propose a multidimentional time-point model and algorithm to solve multi-event expert query parametric estimation (ME-EQPE) problems over multivariate time series. Our proposed model and algorithm combine the strengths of both domain-knowledge-based and formal-learning-based approaches to learn optimal decision parameters for maximising utility over multivariate time series. More specifically, our approach solves the decision optimisation problems to maximise the utility from multiple decision time points, as well as maintaining an optimality of the learned multiple sets of decision parameters in their respective events during the computations. We show that our approach produces a reasonable forecasting result by using the learned multiple sets of decision parameters.

UR - http://www.scopus.com/inward/record.url?scp=84893125195&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893125195&partnerID=8YFLogxK

U2 - 10.1504/IJIDS.2013.055925

DO - 10.1504/IJIDS.2013.055925

M3 - Article

AN - SCOPUS:84893125195

VL - 5

SP - 263

EP - 282

JO - International Journal of Information and Decision Sciences

JF - International Journal of Information and Decision Sciences

SN - 1756-7017

IS - 3

ER -