scholarly journals Nonstationary time series prediction combined with slow feature analysis

2015 ◽  
Vol 2 (1) ◽  
pp. 97-114
Author(s):  
G. Wang ◽  
X. Chen

Abstract. Almost all climate time series have some degree of nonstationarity due to external driving forces perturbations of the observed system. Therefore, these external driving forces should be taken into account when reconstructing the climate dynamics. This paper presents a new technique of combining the driving force of a time series obtained using the Slow Feature Analysis (SFA) approach, then introducing the driving force into a predictive model to predict non-stationary time series. In essence, the main idea of the technique is to consider the driving forces as state variables and incorporate them into the prediction model. To test the method, experiments using a modified logistic time series and winter ozone data in Arosa, Switzerland, were conducted. The results showed improved and effective prediction skill.

2015 ◽  
Vol 22 (4) ◽  
pp. 377-382 ◽  
Author(s):  
G. Wang ◽  
X. Chen

Abstract. Almost all climate time series have some degree of nonstationarity due to external driving forces perturbing the observed system. Therefore, these external driving forces should be taken into account when constructing the climate dynamics. This paper presents a new technique of obtaining the driving forces of a time series from the slow feature analysis (SFA) approach, and then introduces them into a predictive model to predict nonstationary time series. The basic theory of the technique is to consider the driving forces as state variables and to incorporate them into the predictive model. Experiments using a modified logistic time series and winter ozone data in Arosa, Switzerland, were conducted to test the model. The results showed improved prediction skills.


2020 ◽  
Vol 11 (2) ◽  
pp. 525-535
Author(s):  
Xinnong Pan ◽  
Geli Wang ◽  
Peicai Yang ◽  
Jun Wang ◽  
Anastasios A. Tsonis

Abstract. The variations in oceanic and atmospheric modes on various timescales play important roles in generating global and regional climate variability. Many efforts have been devoted to identifying the relationships between the variations in climate modes and regional climate variability, but these have rarely explored the interconnections among these climate modes. Here we use climate indices to represent the variations in major climate modes and examine the harmonic relationship among the driving forces of climate modes using slow feature analysis (SFA) and wavelet analysis. We find that all of the significant peak periods of driving-force signals in the climate indices can be represented as harmonics of four base periods: 2.32, 3.90, 6.55, and 11.02 years. We infer that the period of 2.32 years is associated with the signal of the quasi-biennial oscillation (QBO). The periods of 3.90 and 6.55 years are linked to the intrinsic variability of the El Niño–Southern Oscillation (ENSO), and the period of 11.02 years arises from the sunspot cycle. Results suggest that the base periods and their harmonic oscillations related to QBO, ENSO, and solar activities act as key connections among the climatic modes with synchronous behaviors, highlighting the important roles of these three oscillations in the variability of the Earth's climate. Highlights. i. The harmonic relationship among the driving forces of climate modes was investigated by using slow feature analysis and wavelet analysis.ii. All of the significant peak periods of driving-force signals in climate indices can be represented as the harmonics of four base periods.iii. The four base periods related to QBO, ENSO, and solar activities act as the key linkages among different climatic modes with synchronous behaviors.


2015 ◽  
Vol 124 (3-4) ◽  
pp. 985-989 ◽  
Author(s):  
Geli Wang ◽  
Peicai Yang ◽  
Xiuji Zhou

2012 ◽  
Vol 518-523 ◽  
pp. 5905-5908 ◽  
Author(s):  
Cui Lin Li

Developing geological heritage resource from the perspective of tourism is a new pattern of resource using. The development of Xinjiang geological heritage resource is restricted and effected by the internal and external driving forces and support. The internal driving force includes the resource endowment and the need for regional economic and sustainable development; the external driving force includes the competition of tourism, the market and the government regulation; support includes the construction of infrastructure, the improvement of tourism facilities and ecological support. All these forces restrict and effect each other and become a harmonious whole.


2020 ◽  
Vol 30 (15) ◽  
pp. 2050226
Author(s):  
Yoshito Hirata ◽  
Kazuyuki Aihara

Records for observing dynamics are usually complied by a form of time series. However, time series can be a challenging type of dataset for deep neural networks to learn. In deep neural networks, pairs of inputs and outputs are usually fed for constructive mapping. Such inputs are typically prepared as static images in successful applications. And so, here we propose two methods to prepare such inputs for learning the dynamical properties behind time series. In the first method, we simply array a time series in the shape of a rectangle as an image. In the second method, we convert a time series into a distance matrix using delay coordinates, or an unthresholded recurrence plot. We demonstrate that the second method performs well in inferring a slow driving force from observations of a forced system within which there are symmetry and almost invariant subsets.


2013 ◽  
Vol 34 (1) ◽  
pp. 93-114 ◽  
Author(s):  
Sonia M. Lo

Purpose – The aim of this study is to understand the effect of a firm's position in a supply chain in its industry on the attitude of the firm toward green strategies through empirical data analysis. This study aims to answer the following research questions: Do the environmental uncertainties a firm faces differ with the firm's position in the supply chain when going green?, Would the motivation of a firm for going green vary with uncertainties it faces in the supply chain? and Would green-related practices a firm accepts or executes vary with the firm's position in the supply chain? Design/methodology/approach – The case study method was utilized in this study. The main objects are firms in the high-tech industry of Taiwan, and 12 firms were selected for in-depth investigation. The unit of analysis was a firm. Face-to-face in-depth interviews, approximately 90-105 minutes for each, were conducted with each of the 12 cases. The respondents were middle- to high-level managers. The interviews were recorded and transcribed. Additionally, second-hand information was acquired regarding each case through channels such as firm web sites, documents, and media reports. These integrated data were later utilized in the single-case and cross-case analysis stages. Findings – In this study, firms of Taiwan's high-tech industry are divided into the upstream (raw material supplier), midstream (original design manufacturers/original equipment manufacturer), and downstream (brand company) categories. It is first found the uncertainties a firm encounters when implementing green practices are related to its position in the supply chain. The closer a firm is to the upstream of the supply chain, the higher the competitive uncertainty. In contrast, the closer a firm is to the downstream, the higher the demand uncertainty. Furthermore, the internal and external driving forces of firms in promoting green practices are related to the types of uncertainties the firms encounter in the supply chain. A firm's internal driving force is positively associated with the demand uncertainty it faces, however, negatively with the competition and supply uncertainties. On the other hand, a firm's external driving force is positively associated with the competition and supply uncertainties it faces, however, negatively with the demand uncertainties. Additionally, the association between firm willingness to promote green practices and its position in the supply chain is explored. It is found that, for firms located in the downstream of supply chain, it emphasizes more on the practices of green design, purchase, and internal environmental management. If a firm is located in the midstream of supply chain, it will focus more on the practice of green manufacturing and logistics. Originality/value – This study has expanded the discussion of green supply chain management. It establishes the relationship between the uncertainties and the major driving forces of firms for implementing green practices. This approach is rare in previous literature. Furthermore, past literature has suggested that a specific relationship exists between driving factors and firm practices. The author believes that such a relationship must be based on the position of firms in the supply chain; thus, the author has identified the relationship between supply chain position and green practices.


Sign in / Sign up

Export Citation Format

Share Document