• The Korean Journal of Applied Statistics
• /
• v.22 no.3
• /
• pp.499-513
• /
• 2009

An economic signal in the real world usually reflects complex phenomena. One may have difficulty both extracting and interpreting information embedded in such a signal. A natural way to reduce complexity is to decompose the original signal into several simple components, and then analyze each component. Spectral analysis (Priestley, 1981) provides a tool to analyze such signals under the assumption that the time series is stationary. However when the signal is subject to non-stationary and nonlinear characteristics such as amplitude and frequency modulation along time scale, spectral analysis is not suitable. Huang et al. (1998b, 1999) proposed a data-adaptive decomposition method called empirical mode decomposition and then applied Hilbert spectral analysis to decomposed signals called intrinsic mode function. Huang et al. (1998b, 1999) named this two step procedure the Hilbert-Huang transform(HHT). Because of its robustness in the presence of nonlinearity and non-stationarity, HHT has been used in various fields. In this paper, we discuss the applications of the HHT and demonstrate its promising potential for non-stationary financial time series data provided through a Korean stock price index.

• Communications for Statistical Applications and Methods
• /
• v.30 no.4
• /
• pp.369-388
• /
• 2023

In this paper, we develop the two-step procedure that detects and estimates the position of structural changes for multivariate nonstationary time series, either on mean parameters or second-order structures. We first investigate the presence of mean structural change by monitoring data through the aggregated cumulative sum (CUSUM) type statistic, a sequential procedure identifying the likely position of the change point on its trend. If no mean change point is detected, the proposed method proceeds to scan the second-order structural change by modeling the multivariate nonstationary time series with a multivariate locally stationary Wavelet process, allowing the time-localized auto-correlation and cross-dependence. Under this framework, the estimated dynamic spectral matrices derived from the local wavelet periodogram capture the time-evolving scale-specific auto- and cross-dependence features of data. We then monitor the change point from the lower-dimensional approximated space of the spectral matrices over time by applying the dynamic principal component analysis. Different from existing methods requiring prior information on the type of changes between mean and covariance structures as an input for the implementation, the proposed algorithm provides the output indicating the type of change and the estimated location of its occurrence. The performance of the proposed method is demonstrated in simulations and the analysis of two real finance datasets.

• Journal of Mechanical Science and Technology
• /
• v.20 no.7
• /
• pp.950-960
• /
• 2006

An output time delay always exists in practical systems. Analysis of the delay phenomenon in a continuous-time domain is sophisticated. It is appropriate to obtain its corresponding discrete-time model for implementation via a digital computer. A new method for the discretization of nonlinear systems using Taylor series expansion and the zero-order hold assumption is proposed in this paper. This method is applied to the sampled-data representation of a nonlinear system with a constant output time-delay. In particular, the effect of the time-discretization method on key properties of nonlinear control systems, such as equilibrium properties and asymptotic stability, is examined. In addition, 'hybrid' discretization schemes resulting from a combination of the 'scaling and squaring' technique with the Taylor method are also proposed, especially under conditions of very low sampling rates. A performance of the proposed method is evaluated using two nonlinear systems with time-delay output.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.4B
• /
• pp.355-362
• /
• 2006

Several techniques of MK test, Spearman's Rho test, Linear Regression test, CUSUM test, Cumulative Deviation, Worsley Likelihood Ratio test, Rank Sum test, and Students' t test were applied to detect the trends of slope and shift which exist in hydrologic and climate time series. The time series of annual rainfall, inflow, tree ring index, and southern oscillation index (SOI) were used and the trends of these series were compared in the study. From the results, it can be found that the data could be classified into two categories such as linear trend and shift. 4 series data of 8 rainfall series which reveal the trend show the shift and 8 series data of 18 tree ring index and March and April series of monthly SOI data show shift. Moreover, ADF test and BDS test were used to test stationarity and non-linearity of the data. In conclusion, through the study, various trend analysis techniques were compared and 6 kinds of characteristics which can exist in hydrologic time series were identified.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.43 no.2
• /
• pp.33-38
• /
• 2020

Anomaly detection of Machine Learning such as PCA anomaly detection and CNN image classification has been focused on cross-sectional data. In this paper, two approaches has been suggested to apply ML techniques for identifying the failure time of big time series data. PCA anomaly detection to identify time rows as normal or abnormal was suggested by converting subjects identification problem to time domain. CNN image classification was suggested to identify the failure time by re-structuring of time series data, which computed the correlation matrix of one minute data and converted to tiff image format. Also, LASSO, one of feature selection methods, was applied to select the most affecting variables which could identify the failure status. For the empirical study, time series data was collected in seconds from a power generator of 214 components for 25 minutes including 20 minutes before the failure time. The failure time was predicted and detected 9 minutes 17 seconds before the failure time by PCA anomaly detection, but was not detected by the combination of LASSO and PCA because the target variable was binary variable which was assigned on the base of the failure time. CNN image classification with the train data of 10 normal status image and 5 failure status images detected just one minute before.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.2B
• /
• pp.163-171
• /
• 2009

Classical linear models have been generally used to analyze and forecast hydrologic time series. However, there is growing evidence of nonlinear structure in natural phenomena and hydrologic time series associated with their patterns and fluctuations. Therefore, the classical linear techniques for time series analysis and forecasting may not be appropriate for nonlinear processes. In recent, the BDS (Brock-Dechert-Scheinkman) statistic instead of conventional techniques has been used for detecting nonlinearity of time series. The BDS statistic was derived from the statistical properties of the correlation integral which is used to analyze chaotic system and has been effectively used for distinguishing nonlinear structure in dynamic system from random structures. DVS (Deterministic Versus Stochastic) algorithm has been used for detecting chaos and stochastic systems and for forecasting of chaotic system. This study showed the DVS algorithm can be also used for detecting nonlinearity of the time series. In this study, the stochastic and hydrologic time series are analyzed to detect their nonlinearity. The linear and nonlinear stochastic time series generated from ARMA and TAR (Threshold Auto Regressive) models, a daily streamflow at St. Johns river near Cocoa, Florida, USA and Great Salt Lake Volume (GSL) data, Utah, USA are analyzed, daily inflow series of Soyang dam and the results are compared. The results showed the BDS statistic is a powerful tool for distinguishing between linearity and nonlinearity of the time series and DVS plot can be also effectively used for distinguishing the nonlinearity of the time series.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.2
• /
• pp.321-332
• /
• 2013

Because of the continuous low birthrates, the number of the elementary students will decrease by 17% in 2020 compared to 2011. The purpose of this study is to forecast the number of elementary school teachers until 2020. We used the data in education statistical year books from 1970 to 2010. We used the time-series regression model, time series grouped regression model and exponential smoothing model to predict the number of teachers for the next ten years. Consequently time-series grouped regression model is a better model for forecasting the number of elementary school teachers than other models.

• Proceedings of the CALSEC Conference
• /
• 2005.11a
• /
• pp.147-153
• /
• 2005

When a customer wants to buy an item at the Internet shopping mall, one of the difficulties is to decide when to buy the item because its price changes over time. If the shopping mall can be able to recommend appropriate buying points, it will be greatly helpful for the customer. Therefore, in this presentation, we propose a method to recommend buying points based on the time series analysis using a database that contains past prices data of items. The procedure to provide buying points for an item is as follows. First, we search past time series patterns from the database using normalized similarity, which are similar to the current time series pattern of the item. Second, we analyze the retrieved past patterns and predict the future price pattern of the item. Third, using the future price pattern, we recommend when to buy the item.

• Journal of Information Processing Systems
• /
• v.19 no.2
• /
• pp.149-163
• /
• 2023

Affected by external factors, errors in time series data collected by sensors are common. Using the traditional method of constraining the speed change rate to clean the errors can get good performance. However, they are only limited to the data of stable changing speed because of fixed constraint rules. Actually, data with uneven changing speed is common in practice. To solve this problem, an online cleaning algorithm for time series data based on dynamic speed change rate constraints is proposed in this paper. Since time series data usually changes periodically, we use the extreme learning machine to learn the law of speed changes from past data and predict the speed ranges that change over time to detect the data. In order to realize online data repair, a dual-window mechanism is proposed to transform the global optimal into the local optimal, and the traditional minimum change principle and median theorem are applied in the selection of the repair strategy. Aiming at the problem that the repair method based on the minimum change principle cannot correct consecutive abnormal points, through quantitative analysis, it is believed that the repair strategy should be the boundary of the repair candidate set. The experimental results obtained on the dataset show that the method proposed in this paper can get a better repair effect.

• The Korean Journal of Applied Statistics
• /
• v.24 no.4
• /
• pp.597-607
• /
• 2011

Value at Risk(VaR) is being widely used as a simple tool for measuring financial risk. Although VaR has a few weak points, it is used as a basic risk measure due to its simplicity and easiness of understanding. However, it becomes very difficult to estimate the volatility of the portfolio (essential to compute its VaR) when the number of assets in the portfolio is large. In this case, we can consider the application of a dimension reduction technique; however, the ordinary factor analysis cannot be applied directly to financial data due to autocorrelation. In this paper, we suggest a dimension reduction method that uses the time-series factor analysis and DCC(Dynamic Conditional Correlation) GARCH model. We also compare the method using time-series factor analysis with the existing method using ordinary factor analysis by backtesting the VaR of real data from the Korean stock market.