Dimensionality reduction for large spatio-temporal datasets based on SVD

Sampson, Pd; Guttorp, P

Risultato della ricerca: Paper

Abstract

Many models for spatio-temporal measurements Z(s; t) can be written as a sum of a systematic component and a residual component: Z = M + E. The approach presented here incorporates two Singular Value Decompositions (SVD). The first SVD is applied to the space-time data matrix Z with cross-validation to choose the number of smoothed singular vectors to use as temporal basis functions for modelling spatially varying temporal trend in the matrix M. The second SVD is applied to the spatio-temporal matrix E of residuals from the trend models fitted at each site; it represents spatially correlated short time scale temporal processes. The remaining stochastic structure is explained by simple autoregressive models fit to the final residuals. The procedure is applied to 30 years of daily temperature data from Sicily.
Lingua originaleEnglish
Stato di pubblicazionePublished - 2009

Fingerprint

decomposition
matrix
timescale
modeling
temperature
trend

Cita questo

Dimensionality reduction for large spatio-temporal datasets based on SVD. / Sampson, Pd; Guttorp, P.

2009.

Risultato della ricerca: Paper

Sampson, Pd; Guttorp, P 2009, 'Dimensionality reduction for large spatio-temporal datasets based on SVD'.
Sampson, Pd; Guttorp, P. Dimensionality reduction for large spatio-temporal datasets based on SVD. 2009.
Sampson, Pd; Guttorp, P. / Dimensionality reduction for large spatio-temporal datasets based on SVD.
@conference{5ea803c3770d431f918b3e824a7ad255,
title = "Dimensionality reduction for large spatio-temporal datasets based on SVD",
abstract = "Many models for spatio-temporal measurements Z(s; t) can be written as a sum of a systematic component and a residual component: Z = M + E. The approach presented here incorporates two Singular Value Decompositions (SVD). The first SVD is applied to the space-time data matrix Z with cross-validation to choose the number of smoothed singular vectors to use as temporal basis functions for modelling spatially varying temporal trend in the matrix M. The second SVD is applied to the spatio-temporal matrix E of residuals from the trend models fitted at each site; it represents spatially correlated short time scale temporal processes. The remaining stochastic structure is explained by simple autoregressive models fit to the final residuals. The procedure is applied to 30 years of daily temperature data from Sicily.",
keywords = "Spatio-temporal processes, SVD",
author = "{Sampson, Pd; Guttorp, P} and Rossella Onorati",
year = "2009",
language = "English",

}

TY - CONF

T1 - Dimensionality reduction for large spatio-temporal datasets based on SVD

AU - Sampson, Pd; Guttorp, P

AU - Onorati, Rossella

PY - 2009

Y1 - 2009

N2 - Many models for spatio-temporal measurements Z(s; t) can be written as a sum of a systematic component and a residual component: Z = M + E. The approach presented here incorporates two Singular Value Decompositions (SVD). The first SVD is applied to the space-time data matrix Z with cross-validation to choose the number of smoothed singular vectors to use as temporal basis functions for modelling spatially varying temporal trend in the matrix M. The second SVD is applied to the spatio-temporal matrix E of residuals from the trend models fitted at each site; it represents spatially correlated short time scale temporal processes. The remaining stochastic structure is explained by simple autoregressive models fit to the final residuals. The procedure is applied to 30 years of daily temperature data from Sicily.

AB - Many models for spatio-temporal measurements Z(s; t) can be written as a sum of a systematic component and a residual component: Z = M + E. The approach presented here incorporates two Singular Value Decompositions (SVD). The first SVD is applied to the space-time data matrix Z with cross-validation to choose the number of smoothed singular vectors to use as temporal basis functions for modelling spatially varying temporal trend in the matrix M. The second SVD is applied to the spatio-temporal matrix E of residuals from the trend models fitted at each site; it represents spatially correlated short time scale temporal processes. The remaining stochastic structure is explained by simple autoregressive models fit to the final residuals. The procedure is applied to 30 years of daily temperature data from Sicily.

KW - Spatio-temporal processes, SVD

UR - http://hdl.handle.net/10447/51878

M3 - Paper

ER -

Accesso al documento