scholarly journals Forecasting Development of Economic Processes using Adapted Nonlinear Dynamics Methods

In this work authors propose using adapted nonlinear dynamics methods to prepare time series data for the forecast procedure in order to identify chaotic dynamics and to select forecast methods and models. Each step of the proposed set of methods for data preprocessing allows us to put forward proposals on certain properties of the studied time series. This, in turn, proves that to obtain reliable and reasonable conclusions about the type of behavior of the investigated system, the results of one of the many existing tests are not enough. Conducting a comprehensive analysis, will most correctly determine the type of behavior of the time series and its characteristics, which will make it possible to obtain a reliable forecast in the future.

2021 ◽  
Vol 5 (5) ◽  
pp. 619-635
Author(s):  
Harya Widiputra

The primary factor that contributes to the transmission of COVID-19 infection is human mobility. Positive instances added on a daily basis have a substantial positive association with the pace of human mobility, and the reverse is true. Thus, having the ability to predict human mobility trend during a pandemic is critical for policymakers to help in decreasing the rate of transmission in the future. In this regard, one approach that is commonly used for time-series data prediction is to build an ensemble with the aim of getting the best performance. However, building an ensemble often causes the performance of the model to decrease, due to the increasing number of parameters that are not being optimized properly. Consequently, the purpose of this study is to develop and evaluate a deep learning ensemble model, which is optimized using a genetic algorithm (GA) that incorporates a convolutional neural network (CNN) and a long short-term memory (LSTM). A CNN is used to conduct feature extraction from mobility time-series data, while an LSTM is used to do mobility prediction. The parameters of both layers are adjusted using GA. As a result of the experiments conducted with data from the Google Community Mobility Reports in Indonesia that ranges from the beginning of February 2020 to the end of December 2020, the GA-Optimized Multivariate CNN-LSTM ensemble outperforms stand-alone CNN and LSTM models, as well as the non-optimized CNN-LSTM model, in terms of predicting human movement in the future. This may be useful in assisting policymakers in anticipating future human mobility trends. Doi: 10.28991/esj-2021-01300 Full Text: PDF


2021 ◽  
Vol 12 ◽  
Author(s):  
Suran Liu ◽  
Yujie You ◽  
Zhaoqi Tong ◽  
Le Zhang

It is very important for systems biologists to predict the state of the multi-omics time series for disease occurrence and health detection. However, it is difficult to make the prediction due to the high-dimensional, nonlinear and noisy characteristics of the multi-omics time series data. For this reason, this study innovatively proposes an Embedding, Koopman and Autoencoder technologies-based multi-omics time series predictive model (EKATP) to predict the future state of a high-dimensional nonlinear multi-omics time series. We evaluate this EKATP by using a genomics time series with chaotic behavior, a proteomics time series with oscillating behavior and a metabolomics time series with flow behavior. The computational experiments demonstrate that our proposed EKATP can substantially improve the accuracy, robustness and generalizability to predict the future state of a time series for multi-omics data.


Author(s):  
Mihai Dupac ◽  
Dan B. Marghitu ◽  
David G. Beale

Abstract In this paper, a nonlinear dynamics analysis of the simulated data was considered to study the time evolution of an electro-magnetically levitated flexible droplet. The main goals of this work are to study the behavior of the levitated droplet and to investigate its stability. Quantities characterizing time series data such as attractor dimension or largest Lyapunov exponent were computed.


2007 ◽  
Vol 17 (01) ◽  
pp. 169-182 ◽  
Author(s):  
SIMONE GIANNERINI ◽  
RODOLFO ROSA ◽  
DIEGO LUIS GONZALEZ

The present paper is devoted to the problem of detecting the presence of two positive Lyapunov exponents in time series data. In order to accomplish this task the accuracy of the estimates is essential, but existing estimation approaches do not provide it. We present a procedure exploiting resampling methods for building a statistical test for the presence of two positive exponents of comparable magnitudes through rigorous assessment of confidence intervals. The problem is studied by means of computer experiments performed in a variety of conditions on coupled Lorenz systems. Then, a case study regarding the time series of the cardiovascular activity of the toad Bufo Arenarum is presented. A comparison with other estimator algorithms is also shown.


2020 ◽  
Vol 17 (8) ◽  
pp. 3798-3803
Author(s):  
M. D. Anto Praveena ◽  
B. Bharathi

Big Data analytics has become an upward field, and it plays a pivotal role in Healthcare and research practices. Big data analytics in healthcare cover vast numbers of dynamic heterogeneous data integration and analysis. Medical records of patients include several data including medical conditions, medications and test findings. One of the major challenges of analytics and prediction in healthcare is data preprocessing. In data preprocessing the outlier identification and correction is the important challenge. Outliers are exciting values that deviates from other values of the attribute; they may simply experimental errors or novelty. Outlier identification is the method of identifying data objects with somewhat different behaviors than expectations. Detecting outliers in time series data is different from normal data. Time series data are the data that are in a series of certain time periods. This kind of data are identified and cleared to bring the quality dataset. In this proposed work a hybrid outlier detection algorithm extended LSTM-GAN is helped to recognize the outliers in time series data. The outcome of the proposed extended algorithm attained better enactment in the time series analysis on ECG dataset processing compared with traditional methodologies.


2000 ◽  
Vol 10 (08) ◽  
pp. 1973-1979 ◽  
Author(s):  
TAKAYA MIYANO ◽  
AKIRA NAGAMI ◽  
ISAO TOKUDA ◽  
KAZUYUKI AIHARA

Nonlinear determinism in voiced sounds of the Japanese vowel /a/ is tested by the time series analysis associated with the surrogate method. To capture nonlinear dynamics underlying the speech signal, we apply the generalized radial basis function networks as nonlinear predictors to the time series data. The optimized network parameters may show a trail of the nonlinear dynamics though not conspicuously. This may be due to paucity of data points.


Data ◽  
2021 ◽  
Vol 6 (6) ◽  
pp. 55
Author(s):  
Giuseppe Ciaburro ◽  
Gino Iannace

To predict the future behavior of a system, we can exploit the information collected in the past, trying to identify recurring structures in what happened to predict what could happen, if the same structures repeat themselves in the future as well. A time series represents a time sequence of numerical values observed in the past at a measurable variable. The values are sampled at equidistant time intervals, according to an appropriate granular frequency, such as the day, week, or month, and measured according to physical units of measurement. In machine learning-based algorithms, the information underlying the knowledge is extracted from the data themselves, which are explored and analyzed in search of recurring patterns or to discover hidden causal associations or relationships. The prediction model extracts knowledge through an inductive process: the input is the data and, possibly, a first example of the expected output, the machine will then learn the algorithm to follow to obtain the same result. This paper reviews the most recent work that has used machine learning-based techniques to extract knowledge from time series data.


Sign in / Sign up

Export Citation Format

Share Document