Smoothing splines for trend estimation and prediction in time series
โ Scribed by Richard Morton; Emily L. Kang; Brent L. Henderson
- Publisher
- John Wiley and Sons
- Year
- 2009
- Tongue
- English
- Weight
- 302 KB
- Volume
- 20
- Category
- Article
- ISSN
- 1180-4009
- DOI
- 10.1002/env.925
No coin nor oath required. For personal study only.
โฆ Synopsis
Abstract
We consider the use of generalized additive models with correlated errors for analysing trends in time series. The trend is represented as a smoothing spline so that it can be extrapolated. A method is proposed for choosing the smoothing parameter. It is based on the ability to predict a short term into the future. The choice not only addresses the purpose in hand, but also performs very well, and avoids the tendency to underโsmooth or to interpolate the data that can occur with other dataโdriven methods used to choose the smoothing parameter. The method is applied to data from a chemical process and to stream salinity measurements. Copyright ยฉ 2008 John Wiley & Sons, Ltd.
๐ SIMILAR VOLUMES
## Abstract Let {__X__~__t__~} be a stationary process with spectral density __g__(ฮป).It is often that the true structure __g__(ฮป) is not completely specified. This paper discusses the problem of misspecified prediction when a conjectured spectral density __f__~ฮธ~(ฮป), ฮธโฮ, is fitted to __g__(ฮป). Th
## Abstract Shift trend and step changes were detected for runoff time series in the Shiyang River basin, one of the inland river basins in northโwest China. Annual runoff data from eight tributaries as well as both annual and monthly runoff from the mainstream from 1958 to 2003 were used. Seven st