scholarly journals Modeling Seasonal Variations in Vertical GPS Coordinate Time Series Using Independent Component Analysis and Varying Coefficient Regression

Sensors ◽  
2020 ◽  
Vol 20 (19) ◽  
pp. 5627
Author(s):  
Bin Liu ◽  
Xuemin Xing ◽  
Jianbo Tan ◽  
Qing Xia
Keyword(s):  
Time Series ◽  
Independent Component Analysis ◽  
Seasonal Variations ◽  
Component Analysis ◽  
Independent Component ◽  
Coordinate Time ◽  
Coordinate Time Series ◽  
Varying Coefficient ◽  
The Common ◽  
Gps Coordinate Time Series

Common seasonal variations in Global Positioning System (GPS) coordinate time series always exist, and the modeling and correction of the seasonal signals are helpful for many geodetic studies using GPS observations. A spatiotemporal model was proposed to model the common seasonal variations in vertical GPS coordinate time series, based on independent component analysis and varying coefficient regression method. In the model, independent component analysis (ICA) is used to separate the common seasonal signals in the vertical GPS coordinate time series. Considering that the periodic signals in GPS coordinate time series change with time, a varying coefficient regression method is used to fit the separated independent components. The spatiotemporal model was then used to fit the vertical GPS coordinate time series of 262 global International GPS Service for Geodynamics (IGS) GPS sites. The results show that compared with least squares regression, the varying coefficient method can achieve a more reliable fitting result for the seasonal variation of the separated independent components. The proposed method can accurately model the common seasonal variations in the vertical GPS coordinate time series, with an average root mean square (RMS) reduction of 41.6% after the model correction.

Outliers Detection in Multivariate Time Series by Independent Component Analysis

2007 ◽  
Vol 19 (7) ◽  
pp. 1962-1984 ◽  
Author(s):  
Roberto Baragona ◽  
Francesco Battaglia
Keyword(s):  
Time Series ◽  
Independent Component Analysis ◽  
Model Building ◽  
Multivariate Time Series ◽  
Component Analysis ◽  
Independent Component ◽  
Level Shifts ◽  
The Common ◽  
Identification Techniques ◽  
Pattern Occurrences

In multivariate time series, outlying data may be often observed that do not fit the common pattern. Occurrences of outliers are unpredictable events that may severely distort the analysis of the multivariate time series. For instance, model building, seasonality assessment, and forecasting may be seriously affected by undetected outliers. The structure dependence of the multivariate time series gives rise to the well-known smearing and masking phenomena that prevent using most outliers' identification techniques. It may be noticed, however, that a convenient way for representing multiple outliers consists of superimposing a deterministic disturbance to a gaussian multivariate time series. Then outliers may be modeled as nongaussian time series components. Independent component analysis is a recently developed tool that is likely to be able to extract possible outlier patterns. In practice, independent component analysis may be used to analyze multivariate observable time series and separate regular and outlying unobservable components. In the factor models framework too, it is shown that independent component analysis is a useful tool for detection of outliers in multivariate time series. Some algorithms that perform independent component analysis are compared. It has been found that all algorithms are effective in detecting various types of outliers, such as patches, level shifts, and isolated outliers, even at the beginning or the end of the stretch of observations. Also, there is no appreciable difference in the ability of different algorithms to display the outlying observations pattern.

Sign in / Sign up

Export Citation Format

Share Document