• Korean Management Science Review
• /
• v.8 no.1
• /
• pp.3-11
• /
• 1991

Non-Stationary transition probabilities models which is incorporated into a Markov framework with exogenous variables to account for some of variability are discussed, and extended for alternative procedure. Also as an application of the methodology, the size change of aggregate time-series data on the planted system of agricultural crops is estimated, and evaluated for the precision of time-varying evolution statistically.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.631-634
• /
• 2004

In this paper, the Bayesian recurrent neural network (BRNN) is proposed to predict time series data. Among the various traditional prediction methodologies, a neural network method is considered to be more effective in case of non-linear and non-stationary time series data. A neural network predictor requests proper learning strategy to adjust the network weights, and one need to prepare for non-linear and non-stationary evolution of network weights. The Bayesian neural network in this paper estimates not the single set of weights but the probability distributions of weights. In other words, we sets the weight vector as a state vector of state space method, and estimates its probability distributions in accordance with the Bayesian inference. This approach makes it possible to obtain more exact estimation of the weights. Moreover, in the aspect of network architecture, it is known that the recurrent feedback structure is superior to the feedforward structure for the problem of time series prediction. Therefore, the recurrent network with Bayesian inference, what we call BRNN, is expected to show higher performance than the normal neural network. To verify the performance of the proposed method, the time series data are numerically generated and a neural network predictor is applied on it. As a result, BRNN is proved to show better prediction result than common feedforward Bayesian neural network.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.38 no.2
• /
• pp.65-73
• /
• 2013

In the simulation output analysis, bootstrap method is an applicable resampling technique to insufficient data which are not significant statistically. The moving block bootstrap, the stationary bootstrap, and the threshold bootstrap are typical bootstrap methods to be used for autocorrelated time series data. They are nonparametric methods for stationary time series data, which correctly describe the original data. In the simulation output analysis, however, we may not use them because of the non-stationarity in the data set caused by the trend such as increasing or decreasing. In these cases, we can get rid of the trend by differencing the data, which guarantees the stationarity. We can get the bootstrapped data from the differenced stationary data. Taking a reverse transform to the bootstrapped data, finally, we get the pseudo-samples for the original data. In this paper, we introduce the applicability of bootstrap methods to the time series data having trend, and then verify it through the statistical analyses.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.2
• /
• pp.179-190
• /
• 2017

Hand Accelerometers are widely used to detect human motion patterns in real-time. It is essential to reliably identify which type of activity is performed by human subjects. This rests on having accurate template of each activity. Many human activities are represented as a set of multiple time-series data from such sensors, which are mostly non-stationary and non-linear in nature. This requires a method which can effectively extract patterns from non-stationary and non-linear data. To achieve such a goal, we propose the method to apply Hilbert-Huang Transform which is known to be an effective way of extracting non-stationary and non-linear components from time-series data. It is applied on samples of accelerometer data to determine its effectiveness.

• Journal of Labour Economics
• /
• v.34 no.3
• /
• pp.59-80
• /
• 2011

Using methods for the analysis of non-stationary series and error correction models, this study estimates the elasticity of employment with regard to growth rate and its variability. The authors could not find any significant evidence that elasticity has reduced during the last 25 years, which implies that the slow-down of the employment growth did not result from a reduction of the elasticity but from the GDP growth slow-down. Therefore, policy efforts to enhance the capacity of job creation should be made in a manner that can extend the long-term growth potential.

• Proceedings of the KIEE Conference
• /
• 2000.11d
• /
• pp.666-668
• /
• 2000

In this paper, a fuzzy prediction method is proposed for time series data having uncertainty and non-stationary characteristics. Conventional methods, which use past data directly in prediction procedure, cannot properly handle non-stationary data whose long-term mean is floating. To cope with this problem, a data preprocessing technique utilizing the differences of original time series data is suggested. The difference sets are established from data. And the optimal difference set is selected for input of fuzzy predictor. The proposed method based the Takigi-Sugeno-Kang(TSK or TS) fuzzy rule. Computer simulations show improved results for various time series.

• Journal of the Korean GEO-environmental Society
• /
• v.12 no.4
• /
• pp.53-63
• /
• 2011

Selection or generation of appropriate input ground motion is very important in performing a dynamic analysis. In Korea, it is a common practice to use recorded strong ground motions or artificial motions. The recorded motions show non-stationary characteristics, which is a distinct property of all earthquake motions, but have the problem of not matching the design response spectrum. The artificial motions match the design spectrum, but show stationary characteristics. This study generated ground motions that preserve the non-stationary characteristics of a real earthquake motion, but also matches the design spectrum. In the process, an impulse function-based algorithm that adjusts a given time series in time domain such that it matches the target response spectrum is used. Application of the algorithm showed that it can successfully adjust any recorded motions to match the target spectrum and also preserve the non-stationary characteristics. The modified motions are used to perform a series of nonlinear site response analyses. It is shown that the results using the adjusted motions result in more reliable estimates of ground vibration. It is thus recommended that the newly adjusted motions be used in practice instead of original recorded motions.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.502-507
• /
• 2003

Toward the development of practical methods for observed data oriented bispectral estimation, an automatic means for determining the smoothing bandwidth of bispectral windows is proposed, that can also provide an associated optimum bispectral estimate of stationary non-Gaussian signals, systematically only from an observed time series datum of finite length. For the conventional non-parametric bispectral estimation, the MSE (mean squared error) of the normalized estimate is reviewed under a certain mixing condition and sufficient data length, mainly from the viewpoint of the inverse relation between its bias and variance with respect to the smoothing bandwidth. Based on the fundamental relation, a systematic method not only for determining the bandwidth, but also for obtaining the optimum bispectral estimate is presented by newly introducing a MSE evaluation index of the estimate only from an observed time series datum of finite length. The effectiveness and fundamental features of the proposed method are illustrated by the basic results of numerical experiments.

• The Bulletin of The Korean Astronomical Society
• /
• v.41 no.1
• /
• pp.81.2-81.2
• /
• 2016

Many signals in the nature have power-law behaviors, namely they are "scale-free". The method of detrended fluctuation analysis (DFA), as one of the popular methods (e.g., Rescaled range analysis and Spectral analysis) for determining scale-free nature of time series, has a very important advantage that the DFA can be applied to both stationary and non-stationary signals. The analysis of time series using the DFA has been broadly used in physiology, finance, hydrology, meteorology, geology, and so on. We performed the DFA of 16 Spaceweather HMI Active Region Patch (SHARP) parameters for 38 HMI Active Region Patches (HARPs) obtained by Solar Dynamics Observatory (SDO) from May 2010 to June 2014. The main results from this study are as follows. (1) The most of the time series data are non-stationary. (2) The DFA scaling exponents of "mean vertical current density" for 38 HARPs have a negative correlation coefficient (-0.41) with flare index. (3) The DFA scaling exponents of parameters such as "Sum of the absolute value of net currents per polarity", "Absolute value of the net current helicity", and "Mean photospheric excess magnetic energy density" for the most active HARPs having more than 10 major flares, have positive correlation coefficients (0.64, 0.59, and 0.53, respectively) with the ratio of "the number of CMEs associated with major flares" to "the number of major flares". Physical interpretations on our results will be discussed.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.7 no.2
• /
• pp.301-314
• /
• 2015

Aiming at accurately distinguishing modeless component and natural vibration mode terms from data series of nonlinear and non-stationary processes, such as Vortex-Induced Vibration (VIV), a new empirical mode decomposition method has been developed in this paper. The key innovation related to this technique concerns the method to decompose modeless component from non-stationary process, characterized by a predetermined 'maximum intrinsic time window' and cubic spline. The introduction of conceptual modeless component eliminates the requirement of using spurious harmonics to represent nonlinear and non-stationary signals and then makes subsequent modal identification more accurate and meaningful. It neither slacks the vibration power of natural modes nor aggrandizes spurious energy of modeless component. The scale of the maximum intrinsic time window has been well designed, avoiding energy aliasing in data processing. Finally, it has been applied to analyze data series of vortex-induced vibration processes. Taking advantage of this newly introduced empirical decomposition method and mode identification technique, the vibration analysis about vortex-induced vibration becomes more meaningful.