• 기술혁신연구
• /
• 제27권4호
• /
• pp.21-48
• /
• 2019

기초연구 분야는 정부의 적극적인 지원으로 양적 확대가 큰 폭으로 이루어지는 반면, 체계적인 투자계획이나 데이터에 기반한 재정소요를 제시하는 연구 및 정책자료가 전무하여 관련 연구가 요구되는 시점이다. 이에 본 연구는 시계열 예측모형을 활용하여 기초연구지원사업의 향후 재정소요를 전망하였다. 기초연구분야의 특성을 포함한 다양한 요인들을 종합적으로 고려하기 위하여 시간에 따른 단일 종속변수의 값을 예측하는 ARIMA 모형이 아닌, 다변수의 영향을 반영할 수 있는 ARIMAX 모형을 선택하였다. 모형 적합성 판단을 위해 ARIMAX 모형과 ARIMA 모형의 예측값을 비교한 결과 ARIMAX 모형에서 예측오차율이 개선됨을 확인하였다. ARIMAX 모형에 기반하여 2017년에서 2021년까지 5년 간의 기초연구지원사업 재정소요를 전망하였다. 본 연구는 기초연구지원사업의 재정소요를 통계적 접근방법인 시계열모형을 적용해 전망한 시범적 연구를 수행하였다는 점과, 단변량이 아닌 다변량을 고려하여 예측력을 개선했다는 점에서 의의를 지닌다. 또한 현 정부 국정과제인 '기초연구 예산 2배 확대' 등 기초연구 투자의 중요성이 꾸준히 강조되는 정책기조를 고려할 때 향후 기초연구 투자전략 수립 시 참고자료로 활용 될 수 있다.

• 산업경영시스템학회지
• /
• 제37권2호
• /
• pp.27-34
• /
• 2014

In this paper, we investigate the statistical correlation of the time series for temperature measured at the heat box in the automobile drying process. We show, in terms of the sample variance, that a significant non-linear correlation exists in the time series that consist of absolute temperature changes. To investigate further the non-linear correlation, we utilize the volatility, an important concept in the financial market, and induce volatility time series from absolute temperature changes. We analyze the time series of volatilities in terms of the de-trended fluctuation analysis (DFA), a method especially suitable for testing the long-range correlation of non-stationary data, from the correlation perspective. We uncover that the volatility exhibits a long-range correlation regardless of the window size. We also analyze the cross correlation between two (inlet and outlet) volatility time series to characterize any correlation between the two, and disclose the dependence of the correlation strength on the time lag. These results can contribute as important factors to the modeling of forecasting and management of the heat box's temperature.

• 응용통계연구
• /
• 제31권5호
• /
• pp.667-675
• /
• 2018

초고빈도(ultra high frequency; UHF)시계열의 함수적 변동성 측정을 위한 최신 기법인 함수적 변동성 functional GARCH : fGARCH(1, 1) 모형을 소개하고 설명하였다. 실증분석을 위해 R-code fGARCH(1, 1) 프로그램을 KOSPI/현대차 초고빈도 수익률 자료에 적합하여 예시하였다.

• KDI Journal of Economic Policy
• /
• 제42권1호
• /
• pp.31-56
• /
• 2020

Does financial development contribute to economic growth? The literature finds that an expansion in financial resources is useful for economic growth if the degree of financial development is under a certain threshold; otherwise, the expansion is detrimental to growth. Almost every published study, however, considers country-panel data. Accordingly, the results are not directly applicable to the Korean economy. By examining Korean time-series data, this paper finds that there is an inverse U-shaped relationship between the per capita real GDP growth rate and private credit (as a percentage of nominal GDP)-a well-known measure of quantitative financial development, where the threshold is 171.5%. This paper also finds that private credit is positively associated with economic growth if the share of household credit out of private credit is less than 46.9%; otherwise, private credit is negatively associated with economic growth. As of 2016, the ratio of private credit to GDP and the ratio of household credit to private credit are both higher than the corresponding thresholds, which implies that policymakers should place more emphasis on qualitative financial development than on a quantitative expansion of financial resources.

• 응용통계연구
• /
• 제27권1호
• /
• pp.159-168
• /
• 2014

Engle과 Ng (1993)가 제안한 뉴스 임팩트 커브(NIC)는 표준적인 GARCH 모형에 적용되는 대칭 커브이다. 최근들어 금융시계열의 변동성이 비대칭 성질을 가지는 경향이 있으며 이에 따라 분계점(threshlod) GARCH, 이중선형(bilinear) GARCH 등의 비대칭 모형이 연구되고 있다. 본 논문은 비대칭 모형의 변동성 커브에 대해 연구하고 있으며 LASSO를 통한 방법론을 제안하고 있다. 제시된 방법론을 국내 KOSDAQ 자료분석을 통해 예시해 보았다.

• Communications for Statistical Applications and Methods
• /
• 제15권6호
• /
• pp.825-835
• /
• 2008

계량경제학 분야에서 널리 쓰이는 MGARCH(multivariate GARCH)모형은 여러개의 시계열자료들의 변동성을 함께 모형화한다. 그러나 변수가 많아질수록 추정해야 할 모수의 수가 급격하게 늘어나는 문제점이 있다. 본 연구에서는 인자 모형을 통해 자료의 차원을 축소시킴로써 이러한 문제를 해결하고자 하였다. 국내의 주가수익률 자료에 통계적 인자 모형과 fundamental factor model을 적용하여 각각의 의미 있는 인자들을 얻은 후 이를 MGARCH모형에 적합시켰다. 또한 두 인자모형을 바탕으로 얻어진 최종 모형들의 MSE, MAD와 VaR(Value at Risk)를 계산하여 예측력을 비교하고자 한다.

• 응용통계연구
• /
• 제27권7호
• /
• pp.1219-1228
• /
• 2014

다변량 금융시계열 분석모형인 상수조건부상관(CCC)에 대해 알아보았으며, 개개 변동성간의 상호작용을 함께 고려한 확장된 상수조건부상관(ECCC)을 소개하고 국내 금융시계열에 적용하였다. 다양한 이변량 수익률 자료를 통해 CCC와 ECCC를 비교분석하였다.

• 한국데이타베이스학회:학술대회논문집
• /
• 한국데이타베이스학회 1999년도 춘계공동학술대회: 지식경영과 지식공학
• /
• pp.175-186
• /
• 1999

Detecting the features of significant patterns from their own historical data is so much crucial to good performance specially in time-series forecasting. Recently, a new data filtering method (or multi-scale decomposition) such as wavelet analysis is considered more useful for handling the time-series that contain strong quasi-cyclical components than other methods. The reason is that wavelet analysis theoretically makes much better local information according to different time intervals from the filtered data. Wavelets can process information effectively at different scales. This implies inherent support fer multiresolution analysis, which correlates with time series that exhibit self-similar behavior across different time scales. The specific local properties of wavelets can for example be particularly useful to describe signals with sharp spiky, discontinuous or fractal structure in financial markets based on chaos theory and also allows the removal of noise-dependent high frequencies, while conserving the signal bearing high frequency terms of the signal. To date, the existing studies related to wavelet analysis are increasingly being applied to many different fields. In this study, we focus on several wavelet thresholding criteria or techniques to support multi-signal decomposition methods for financial time series forecasting and apply to forecast Korean Won / U.S. Dollar currency market as a case study. One of the most important problems that has to be solved with the application of the filtering is the correct choice of the filter types and the filter parameters. If the threshold is too small or too large then the wavelet shrinkage estimator will tend to overfit or underfit the data. It is often selected arbitrarily or by adopting a certain theoretical or statistical criteria. Recently, new and versatile techniques have been introduced related to that problem. Our study is to analyze thresholding or filtering methods based on wavelet analysis that use multi-signal decomposition algorithms within the neural network architectures specially in complex financial markets. Secondly, through the comparison with different filtering techniques' results we introduce the present different filtering criteria of wavelet analysis to support the neural network learning optimization and analyze the critical issues related to the optimal filter design problems in wavelet analysis. That is, those issues include finding the optimal filter parameter to extract significant input features for the forecasting model. Finally, from existing theory or experimental viewpoint concerning the criteria of wavelets thresholding parameters we propose the design of the optimal wavelet for representing a given signal useful in forecasting models, specially a well known neural network models.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 1999년도 춘계공동학술대회-지식경영과 지식공학
• /
• pp.175-186
• /
• 1999

Detecting the features of significant patterns from their own historical data is so much crucial to good performance specially in time-series forecasting. Recently, a new data filtering method (or multi-scale decomposition) such as wavelet analysis is considered more useful for handling the time-series that contain strong quasi-cyclical components than other methods. The reason is that wavelet analysis theoretically makes much better local information according to different time intervals from the filtered data. Wavelets can process information effectively at different scales. This implies inherent support for multiresolution analysis, which correlates with time series that exhibit self-similar behavior across different time scales. The specific local properties of wavelets can for example be particularly useful to describe signals with sharp spiky, discontinuous or fractal structure in financial markets based on chaos theory and also allows the removal of noise-dependent high frequencies, while conserving the signal bearing high frequency terms of the signal. To data, the existing studies related to wavelet analysis are increasingly being applied to many different fields. In this study, we focus on several wavelet thresholding criteria or techniques to support multi-signal decomposition methods for financial time series forecasting and apply to forecast Korean Won / U.S. Dollar currency market as a case study. One of the most important problems that has to be solved with the application of the filtering is the correct choice of the filter types and the filter parameters. If the threshold is too small or too large then the wavelet shrinkage estimator will tend to overfit or underfit the data. It is often selected arbitrarily or by adopting a certain theoretical or statistical criteria. Recently, new and versatile techniques have been introduced related to that problem. Our study is to analyze thresholding or filtering methods based on wavelet analysis that use multi-signal decomposition algorithms within the neural network architectures specially in complex financial markets. Secondly, through the comparison with different filtering techniques results we introduce the present different filtering criteria of wavelet analysis to support the neural network learning optimization and analyze the critical issues related to the optimal filter design problems in wavelet analysis. That is, those issues include finding the optimal filter parameter to extract significant input features for the forecasting model. Finally, from existing theory or experimental viewpoint concerning the criteria of wavelets thresholding parameters we propose the design of the optimal wavelet for representing a given signal useful in forecasting models, specially a well known neural network models.

• 한국항해항만학회지
• /
• 제40권2호
• /
• pp.57-65
• /
• 2016

The port throughput situation has changed since the 2008 financial crisis in the US. Therefore, we studied the situation, accurately estimating port traffic of Korean port after the 2008 financial crisis. We ensured the proper port facilities in response to changes in port traffic. In the results of regression analysis, Korean GDP and the real effective exchange rate of Korean Won were found to increase the container throughput in Korean and Busan port, as well as trade volume with China. Also, the real effective exchange rate of Korean Won was found to increase the port transshipment cargo volume. Based on the ARIMA models, we forecasted port throughput and port transshipment cargo volume for the next six years (72 months), from 2015 to 2020. As a result, port throughput of Korean and Busan ports was forecasted by increasing annual the average from about 3.5% to 3.9%, and transshipment cargo volume was forecasted by increasing the annual average about 4.5%.