MetaCartSign in to MyCiteSeer

Include Citations | Advanced Search | Help

Include Citations | Advanced Search | Help

  and

Download:
Download as a PDF | Download as a PS
by Guy P. Nason, Theofanis Sapatinas
http://www.ukc.ac.uk/IMS/statistics/people/T.Sapatinas/links/report_ims99-31.ps.gz
Add To MetaCart

Abstract:

This article shows how a non-decimated wavelet packet transform (NWPT) can be used to model a response time series Y t in terms of an explanatory time series X t. We apply our technique to an important problem from the wind energy industry: how to model wind speed at a target location using wind speed and direction from a reference location. Our technique transforms the explanatory series, X t, into a NWPT representation and then uses standard statistical modelling methods to identify which wavelet packets are useful for modelling Y t. Our methods often generate better target site wind speed predictions compared to those produced by industry standard techniques. However, of more importance, our wavelet packet techniques produce models to which we can attach physical and scientic interpretations to understand more about the transfer of wind energy from site to site.

Citations

1556 A theory for multiresolution signal decomposition: the wavelet representation – Mallat - 1989
378 Entropy-based algorithms for best basis selection – Coifman, Wickerhauser - 1992
259 Spectral analysis and time series – Priestley - 1989
216 Adapted Wavelet Analysis from Theory to S(~#ware – Wickerhauser - 1994
156 Introduction to Wavelets and Wavelet Transforms: A Primer – Burrus, Gopinath, et al. - 1998
140 Wavelet thresholding via a Bayesian approach – Abramovich, Sapatinas, et al. - 1998
103 Statistical Modeling by Wavelets – Vidakovic - 1999
89 The stationary wavelet transform and some statistical applications – Nason, Silverman - 1995
83 The discrete wavelet transform in – Nason, Silverman - 1994
62 Space-Time Modeling with Long Memory Dependence: Assessing Ireland’s Wind Power Resource (with discussion – Haslett, Raftery - 1989
60 Fitting time series models to non-stationary processes – Dahlhaus - 1997
32 Time-invariant orthonormal wavelet representations – Pesquet, Krim, et al. - 1996
29 Discriminant analysis with singular covariance matrices: methods and applications to spectroscopic data, Applied Statistics 44 – Krzanowski, Jonathan, et al. - 1995
26 Orthonormal shift-invariant wavelet packet decomposition and representation – Cohen, Raz, et al. - 1997
24 Wavelet processes and adaptive estimation of the evolutionary spectrum – Nason - 2000
20 Wavelets and time-frequency analysis – Hess-Nielsen, Wickerhauser - 1996
12 The designer’s guide to wind loading of building structures part 2 – Cook - 1997
9 The phase-corrected undecimated discrete wavelet packet transform and its application to the timing of events – Walden, Cristan, et al. - 1998
8 Power Spectrum of Horizontal Wind Speed in the Frequency Range from 0.0007 to 900 Cycles per Hour – Hoven - 1957
8 The Decomposition of Economic Relationships by Time Scale Using Wavelets: Expenditure and Income – Ramsey, Lampart - 1998
5 Prediction of extreme wind speeds at wind energy sites – Hannah, Palutikof, et al. - 1996
3 Non-decimated wavelet modelling of infant sleep state in terms of ECG – Nason, Sapatinas, et al. - 1999