• Wind and Structures
• /
• v.31 no.5
• /
• pp.445-457
• /
• 2020

The buffeting response is a vital consideration for long-span bridges in typhoon-prone areas. In the conventional analysis, the turbulence and structural vibrations are assumed as stationary processes, which are, however, inconsistent with the non-stationary features observed in typhoon winds. This poses a question on how the stationary assumption would affect the evaluation of buffeting responses under non-stationary wind actions in nature. To figure out this problem, this paper presents a comparative study on buffeting responses of a long-span cable-stayed bridge based on stationary and non-stationary perspectives. The stationary and non-stationary buffeting analysis frameworks are firstly reviewed. Then, a modal analysis of the example bridge, Sutong Cable-stayed Bridge (SCB), is conducted, and stationary and non-stationary spectral models are derived based on measured typhoon winds. On this condition, the buffeting responses of SCB are finally analyzed by following stationary and non-stationary approaches. Although the stationary results are almost identical with the non-stationary results in the mean sense, the root-mean-square value of buffeting responses are underestimated by the stationary assumption as the time-varying features existing in the spectra of turbulence are neglected. The analytical results highlights a transition from stationarity to non-stationarity in the buffeting analysis of long-span bridges.

• 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.17 no.2
• /
• pp.241-253
• /
• 2010

As standard unit root tests are empirically proved to fail to reject the null hypothesis of a unit root for many economic and business time series, it is doubtful that most of those series are informative about the existence of a unit root or that those tests are powerful against relevant alternative hypotheses. This study attempts to perform tests of the null hypothesis of stationarity as well as tests of the null hypothesis of a unit root using the time series data of housing prices in the major metropolitan areas. The results of the additional analyses such as lead-lag, cross-correlation and impulse response for testing the statistical interrelationships between the prices are generally found to be consistent.

• Wind and Structures
• /
• v.32 no.2
• /
• pp.89-104
• /
• 2021

Non-synoptic winds generated by tornadoes, downbursts or gust fronts exhibit significant non-stationarity and can cause significant wind load effect on flexible structures such as long-span bridges. However, conventional assumptions on stationarity used to evaluate the structural wind-induced vibration are inadequate. In this paper, an efficient frequency domain scheme based on fast CQC method, which can predict non-stationary buffeting random responses of long-span bridges, is presented, and then this approach is applied to evaluate the buffeting response of a long-span suspension bridge located in a complex mountainous wind environment as an example. In this study, the data-driven method based on one available measured wind speed sample is firstly presented to establish non-stationary wind models, including time-varying mean wind speed, time-varying intensity envelope function and uniformly modulated fluctuating spectrum. Then, a linear time-variant (LTV) system based on the proposed scheme can be generally applied to calculate the non-stationary buffeting responses. The effectiveness and accuracy of the proposed scheme are verified through Monte Carlo time domain simulation implemented in ANSYS platform. Also, the transient effect nature of the bridge responses is further illustrated by comparison of the non-stationary, quasistationary and steady-state cases. Finally, buffeting response analysis with traditional stationary treatment (10 min constant mean plus stationary wind fluctuation) is performed to illustrate the importance of the non-stationary characteristics embedded in original wind speed samples.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.57 no.5
• /
• pp.139-152
• /
• 2015

Along with climate change, it is reported that the scale and frequency of extreme climate events show unstable tendency of increase. Thus, to comprehend the change characteristics of precipitation data, it is needed to consider non-stationary. The main objectives of this study were to estimate future design floods for Wonpyeongcheon watershed based on RCP (Representative Concentration Pathways) scenario. Wonpyeongcheon located in the Keum River watershed was selected as the study area. Historical precipitation data of the past 35 years (1976~2010) were collected from the Jeonju meteorological station. Future precipitation data based on RCP4.5 were also obtained for the period of 2011~2100. Systematic bias between observed and simulated data were corrected using the quantile mapping (QM) method. The parameters for the bias-correction were estimated by non-parametric method. A non-stationary frequency analysis was conducted with moving average method which derives change characteristics of generalized extreme value (GEV) distribution parameters. Design floods for different durations and frequencies were estimated using rational formula. As the result, the GEV parameters (location and scale) showed an upward tendency indicating the increase of quantity and fluctuation of an extreme precipitation in the future. The probable rainfall and design flood based on non-stationarity showed higher values than those of stationarity assumption by 1.2%~54.9% and 3.6%~54.9%, respectively, thus empathizing the necessity of non-stationary frequency analysis. The study findings are expected to be used as a basis to analyze the impacts of climate change and to reconsider the future design criteria of Wonpyeongcheon watershed.

• Smart Media Journal
• /
• v.9 no.4
• /
• pp.26-35
• /
• 2020

Attempts to predict future stock prices have been studied steadily since the past. However, unlike general time-series data, financial time-series data has various obstacles to making predictions such as non-stationarity, long-term dependence, and non-linearity. In addition, variables of a wide range of data have limitations in the selection by humans, and the model should be able to automatically extract variables well. In this paper, we propose a 'sliding time step normalization' method that can normalize non-stationary data and LSTM autoencoder to compress variables from all variables. and 'moving transfer learning', which divides periods and performs transfer learning. In addition, the experiment shows that the performance is superior when using as many variables as possible through the neural network rather than using only 100 major financial variables and by using 'sliding time step normalization' to normalize the non-stationarity of data in all sections, it is shown to be effective in improving performance. 'moving transfer learning' shows that it is effective in improving the performance in long test intervals by evaluating the performance of the model and performing transfer learning in the test interval for each step.

• Honam Mathematical Journal
• /
• v.16 no.1
• /
• pp.111-117
• /
• 1994

In this note we prove a functional central. limit theorem for LPQD sequences, statisfying some moment conditions. No stationarity is required. Our results imply an extension of Birkel's functional central limit theorem for associated processt'S to an LPQD sequence and an improvement of Birkel's functional central limit theorem for LPQD sequences.

• Journal of the Korean Statistical Society
• /
• v.25 no.1
• /
• pp.153-159
• /
• 1996

Sums of independent random variables $S_n = X_1 + X_ + cdots + X_n$ are considered, where the X$_{n}$ are chosen according to a stationary process of distributions. Given the time t .geq. O, let N (t) be the number of indices n for which O < $S_n$ $\geq$ t. In this set up we prove that N (t)/t converges almost surely and in $L^1$ as t longrightarrow $\infty$, which generalizes classical renewal theorem.m.

• Journal of the Korean Statistical Society
• /
• v.21 no.2
• /
• pp.127-137
• /
• 1992

In this paper we investigate scaling limits for associated random measures satisfying some moment conditions. No stationarity is required. Our results imply an improvement of a central limit theorem of Cox and Grimmett to associated random measure and an extension to the nonstationary case of scaling limits of Burton and Waymire. Also we prove an invariance principle for associated random measures which is an extension of the Birkel's invariance principle for associated process.

• Communications for Statistical Applications and Methods
• /
• v.17 no.5
• /
• pp.639-645
• /
• 2010

We consider a subdiagonal bilinear model and give sufficient conditions for the associated Markov chain defined by Pham (1985) to be uniformly ergodic and then obtain the $\beta$-mixing property for the given process. To derive the desired properties, we employ the results of generalized random coefficient autoregressive models generated by a matrix-valued polynomial function and vector-valued polynomial function.