Home > Research > Publications & Outputs > Environmental Time Series Analysis and Forecast...
View graph of relations

Environmental Time Series Analysis and Forecasting with the Captain Toolbox

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Environmental Time Series Analysis and Forecasting with the Captain Toolbox. / Taylor, C. James; Pedregal, Diego J.; Young, Peter C.; Tych, Wlodek.

In: Environmental Modelling and Software, Vol. 22, No. 6, 01.06.2007, p. 797-814.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Author

Taylor, C. James ; Pedregal, Diego J. ; Young, Peter C. ; Tych, Wlodek. / Environmental Time Series Analysis and Forecasting with the Captain Toolbox. In: Environmental Modelling and Software. 2007 ; Vol. 22, No. 6. pp. 797-814.

Bibtex

@article{dc7583d99be848f1b8ebef518439e706,
title = "Environmental Time Series Analysis and Forecasting with the Captain Toolbox",
abstract = "The Data-Based Mechanistic (DBM) modelling philosophy emphasises the importance of parametrically efficient, low order, {\textquoteleft}dominant mode{\textquoteright} models, as well as the development of stochastic methods and the associated statistical analysis required for their identification and estimation. Furthermore, it stresses the importance of explicitly acknowledging the basic uncertainty in the process, which is particularly important for the characterisation and forecasting of environmental and other poorly defined systems. The paper focuses on a Matlab{\textregistered} compatible toolbox that has evolved from this DBM modelling research. Based around a state space and transfer function estimation framework, Captain extends Matlab{\textregistered} to allow, in the most general case, for the identification and estimation of a wide range of unobserved components models. Uniquely, however, Captain focuses on models with both time variable and state dependent parameters and has recently been implemented with the latest methodological developments in this regard. Here, the main innovations are: the automatic optimisation of the hyper-parameters, which define the statistical properties of the time variable parameters; the provision of smoothed as well as filtered parameter estimates; the robust and statistically efficient identification and estimation of both discrete and continuous time transfer function models; and the availability of various special model structures that have wide application potential in the environmental sciences.",
keywords = "Data-based mechanistic, Identification, Forecasting, Signal processing, Unobserved components model, Kalman filtering, Fixed interval smoothing, Hyper-parameter optimisation, Maximum likelihood",
author = "Taylor, {C. James} and Pedregal, {Diego J.} and Young, {Peter C.} and Wlodek Tych",
year = "2007",
month = jun,
day = "1",
doi = "10.1016/j.envsoft.2006.03.002",
language = "English",
volume = "22",
pages = "797--814",
journal = "Environmental Modelling and Software",
issn = "1364-8152",
publisher = "Elsevier BV",
number = "6",

}

RIS

TY - JOUR

T1 - Environmental Time Series Analysis and Forecasting with the Captain Toolbox

AU - Taylor, C. James

AU - Pedregal, Diego J.

AU - Young, Peter C.

AU - Tych, Wlodek

PY - 2007/6/1

Y1 - 2007/6/1

N2 - The Data-Based Mechanistic (DBM) modelling philosophy emphasises the importance of parametrically efficient, low order, ‘dominant mode’ models, as well as the development of stochastic methods and the associated statistical analysis required for their identification and estimation. Furthermore, it stresses the importance of explicitly acknowledging the basic uncertainty in the process, which is particularly important for the characterisation and forecasting of environmental and other poorly defined systems. The paper focuses on a Matlab® compatible toolbox that has evolved from this DBM modelling research. Based around a state space and transfer function estimation framework, Captain extends Matlab® to allow, in the most general case, for the identification and estimation of a wide range of unobserved components models. Uniquely, however, Captain focuses on models with both time variable and state dependent parameters and has recently been implemented with the latest methodological developments in this regard. Here, the main innovations are: the automatic optimisation of the hyper-parameters, which define the statistical properties of the time variable parameters; the provision of smoothed as well as filtered parameter estimates; the robust and statistically efficient identification and estimation of both discrete and continuous time transfer function models; and the availability of various special model structures that have wide application potential in the environmental sciences.

AB - The Data-Based Mechanistic (DBM) modelling philosophy emphasises the importance of parametrically efficient, low order, ‘dominant mode’ models, as well as the development of stochastic methods and the associated statistical analysis required for their identification and estimation. Furthermore, it stresses the importance of explicitly acknowledging the basic uncertainty in the process, which is particularly important for the characterisation and forecasting of environmental and other poorly defined systems. The paper focuses on a Matlab® compatible toolbox that has evolved from this DBM modelling research. Based around a state space and transfer function estimation framework, Captain extends Matlab® to allow, in the most general case, for the identification and estimation of a wide range of unobserved components models. Uniquely, however, Captain focuses on models with both time variable and state dependent parameters and has recently been implemented with the latest methodological developments in this regard. Here, the main innovations are: the automatic optimisation of the hyper-parameters, which define the statistical properties of the time variable parameters; the provision of smoothed as well as filtered parameter estimates; the robust and statistically efficient identification and estimation of both discrete and continuous time transfer function models; and the availability of various special model structures that have wide application potential in the environmental sciences.

KW - Data-based mechanistic

KW - Identification

KW - Forecasting

KW - Signal processing

KW - Unobserved components model

KW - Kalman filtering

KW - Fixed interval smoothing

KW - Hyper-parameter optimisation

KW - Maximum likelihood

U2 - 10.1016/j.envsoft.2006.03.002

DO - 10.1016/j.envsoft.2006.03.002

M3 - Journal article

VL - 22

SP - 797

EP - 814

JO - Environmental Modelling and Software

JF - Environmental Modelling and Software

SN - 1364-8152

IS - 6

ER -