Trend assessment in a long memory dependence model using the discrete wavelet transform

By Peter F. Craigmile, Peter Guttorp, Donald B. Percival. 2004

68 citations recorded by the source corpus.

Subfields: Applied Mathematics, Computer Vision and Pattern Recognition, Finance

View Paper: https://www.semanticscholar.org/paper/743afa74972255f70fa4893e9ce9943126241805

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Applied Mathematics

In Computer Vision and Pattern Recognition

In Finance

Citations in the corpus

Statistical analysis of hydroclimatic time series: Uncertainty and insights (2007, background): “…series including wind power and indexes of North Atlantic Oscillation [Haslett andRaftery, 1989; Stephenson et al., 2000] as well as proxy series such as tree ring widths or isotope concentrations [Pelletier and Turcotte, 1997; Koutsoyiannis, 2002; Beran and Feng, 2002; Craigmile et al., 2004b].”
Discrete wavelet‐based trend identification in hydrologic time series (2013, background): “Craigmile et al. (2004) took trend to be continuously and slowly varying changes in a time series over long scales.”
1 Trend Filtering (2009, methodology): “including Hodrick–Prescott (H-P) filtering [52, 64], moving average filtering [75], exponential smoothing [70], bandpass filtering [21, 4], smoothing splines [81], de-trending via rational square-wave filters [79], a jump process approach [106], median filtering [101], a linear programming (LP) appr”
Space‐time modelling of trends in temperature series (2011, methodology): “…integer Jl J. Letting 0m denote a vector of zeroes, an estimate of Wl is Ŵl ¼ ð0N1 ; . . . ; 0NJl ;VY;JlÞ T : ð6Þ (An embellishment of this estimate includes wavelet coefficients on lower scales that are affected by filtering at the boundaries – see Craigmile et al. (2004) for further details.)”
Nature's style: Naturally trendy (2005, methodology): “[3] We assume that an HC record, ~ Y � (Y1, ... , YN) 0 , arises from a stochastic process, and that the process can be partitioned into a deterministic linear trend component and a stochastic component [Kendall et al., 1983; Craigmile et al., 2004 ] such that”