• Journal of the Korea Convergence Society
• /
• v.12 no.4
• /
• pp.249-258
• /
• 2021

Bitcoin prices have been soaring recently as investors flock to cryptocurrency exchanges. The purpose of this study is to predict the Bitcoin price using a deep learning model and analyze whether Bitcoin is profitable through investment strategy. LSTM is utilized as Bitcoin prediction model with nonlinearity and long-term memory and the profitability of MA cross-over strategy with predicted prices as input variables is analyzed. Investment performance of Bitcoin strategy using LSTM forecast prices from 2013 to 2021 showed return improvement of 5.5% and 46% more than market price MA cross-over strategy and benchmark Buy & Hold strategy, respectively. The results of this study, which expanded to recent data, supported the inefficiency of the cryptocurrency market, as did previous studies, and showed the feasibility of using the deep learning model for Bitcoin investors. In future research, it is necessary to develop optimal prediction models and improve the profitability of Bitcoin investment strategies through performance comparison of various deep learning models.

• The Journal of Engineering Geology
• /
• v.13 no.4
• /
• pp.457-474
• /
• 2003

This study aims to elucidate characteristics of groundwater level fluctuation at riverbank filtration sites in Daesan-myeon, Changwon City. Groundwater level fluctuation, river water level change and stream-aquifer interaction are very important to estimate optimal discharge rate of the pumping well. Water level contours from February 2003 to October 2003 show normal decreasing trend toward the Nakdong river with the hydraulic gradient of 0.008. However, flow reversion occurs when groundwater is discharged at the pumping wells or rise of the Nakdong river by rainfall. The fluctuation of the Nakdong river ranges 0 - 10 m msl. Autocorrelation analysis was conducted to the groundwater levels measured on the six monitoring wells (DS1, DS2, DS3, DS4, DS6 and DS7). The analyzed waterlevel data can be grouped into three: group 1 (DS1 and DS3) represents strong linearity and long memory effect, group 2 (DS1 and DS6) intermediate linearity and memory, and group 3 (DS4 and DS7) weak linearity and memory. Waterlevels of group 1 wells are relatively closely related to the change of river-water level. Those of group 2 wells are largely affected by the pumping and the river-water level, and those of group 3 wells are strongly linked to pumping.

• Information and Communications Magazine
• /
• v.11 no.2
• /
• pp.113-122
• /
• 1994

In this paper, We analyze and inspect the problem of our country PSTN subscriber telephone number and suggest the ISDN/PSTN subsciber numbering plan in the future. Expecially, when you eastablish plan to change present telephone number system to ISDN telephone number system, public goverment and public office which has many calls have to be given a same number that many people want and being easy rememberance thoughout the country. Therefore we think that it is saving hours and reduce lost calls. As concrete action of convert plan it must be made consistent numbering plan when the numbering reorganization is carried out in Korea before T-time. If a partial is not carried out, it should be given when the telephone number is newly-organized or the big buliding and goverment and public office eastablich RSS individually. At that time chaos will be minimized.

• Journal of Fashion Business
• /
• v.16 no.2
• /
• pp.124-137
• /
• 2012

This study aims to identify the meaning and formation cause of a style, and the essential elements of style formation, through psychobiological research as well as an analysis of the designs of Iris Van Herpen, a fashion designer, who in just 6 years has developed into a world-renowned new designer. As a result, it has been found that the psychobiological causes to form a style stem from the action of 'long-term memory', which is consolidated by 'selective attention', 'perceptional subjectivity', the principle of the 'neuron's connection specificity and invariance', and the principle of a 'neuronal signal's unilateral flow'. With such action, Herpen could develop her own original composition techniques. The formative shapes created by such composition techniques are characterized by enumeration, superposition, and hanging. The study has also found that the essential elements for a designer to be able to form his/her own style include 'aesthetic originality' in which the designer views the property of a thing from his/her inherent perspective, and finds the uniqueness from the thing that only he/she can express, 'technical differences' that are creative and original, and 'formative specificity' that is summarized into one property through an impressive shape.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.366-366
• /
• 2022

유역의 하천유량과 같은 수문 시계열을 모의 또는 예측하기 위한 수문 모델링에서 최근 기계 학습 방법을 활용한 연구가 활발하게 적용되고 있는 추세이다. 이러한 데이터 기반 모델링 접근법은 입출력 자료에서 관찰된 패턴을 학습하며, 특히, 장단기기억(Long Short-Term Memory, LSTM) 네트워크는 많은 연구에서 수문 시계열 예측에 대한 적용성이 검증되었으나, 장기간의 고품질 관측자료를 활용할 때 더 나은 예측성능을 보인다. 그러나 우리나라의 경우 장기간 관측된 고품질의 하천유량 자료를 확보하기 어려운 실정이다. 따라서 본 연구에서는 LSTM 네트워크의 학습 시 가용한 모든 유역의 자료를 통합하여 학습시켰을 때 하천유량 예측성능을 개선할 수 있는지 판단해보고자 하였다. 이를 위해, 우리나라 13개 댐 유역을 대상으로 대상 유역의 자료만을 학습한 모델의 예측성능과 모든 유역의 자료를 학습한 모델의 예측성능을 비교해 보았다. 학습은 2001년부터 2010년까지 기상자료(강우, 최저·최고·평균기온, 상대습도, 이슬점, 풍속, 잠재증발산)를 이용하였으며, 2011년부터 2020년에 대해 테스트 되었다. 다지점 통합학습을 통해 테스트 기간에 대해 예측된 각 유역의 일 하천유량의 KGE 중앙값이 0.74로 단일지점 학습을 통해 예측된 KGE(0.72)보다 다소 개선된 결과를 보여주었다. 다지점 통합학습이 하천유량 예측에 큰 개선을 달성하지는 못하였으며, 추가적인 가용 자료 확보와 LSTM 구성의 개선을 통해 추가적인 연구가 필요할 것으로 판단된다.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.9 no.2
• /
• pp.15-21
• /
• 2014

The long memory properties of the hedge ratio for stock and futures have not been systematically investigated by the extant literature. To investigate hedge ratio' long memory, this paper employs a data set including KOSPI200 and S&P500. Coakley, Dollery, and Kellard(2008) employ a data set including a stock index and commodities foreign exchange, and suggested the S&P500 to be a fractionally integrated process. This paper firstly estimates hedge ratios with two dynamic models, BEKK(Bollerslev, Engle, Kroner, and Kraft) and diagonal-BEKK, and tests the long memory of hedge ratios with Geweke and Porter-Hudak(1983)(henceforth GPH) and Lo's modified rescaled adjusted range test by Lo(1991). In empirical results, two hedge ratios based on KOSPI200 and S&P500 show considerably significant long memory behaviours. Thus, such results show the hedge ratios to be stationary and strongly reject the random walk hypothesis on hedge ratios, which violates the efficient market hypothesis.

• Korean Journal of Food Science and Technology
• /
• v.44 no.4
• /
• pp.428-433
• /
• 2012

Probiotics and their products, such as yogurt and cheese have been widely consumed in many countries with proven health benefits including anti-microbial activity and anti-diarrheal activity. LHFM (Lactobacillus helveticus - fermented milk) is a processed skim milk powder, fermented by a probiotics, L. helveticus IDCC3801. In the present study, we aimed to investigate the neuroprotective effects and the cognitive improvements of LHFM. LHFM itself did not show any cytotoxicity to the human neuroblastoma cell line, SH-SY5Y; however, it dose-dependently protected against glutamate-induced neuronal cell death. LHFM also attenuated scopolamine-induced memory deficit in Y-maze and Morris-water maze. In the analysis of hippocampus after a behavior test, LHFM significantly increased the acetylcholine level and also inhibited acetylcholine esterase activity. Therefore, the raised acetylcholine release partially contributes to the improvement of learning and memory by a treatment with LHFM. These results suggest that LHFM is an effective material for prevention or improvement of cognitive impairments caused by neuronal cell damage and central cholinergic dysfunction.

• Journal of Korea Water Resources Association
• /
• v.52 no.8
• /
• pp.575-587
• /
• 2019

This study evaluated the effect of a mother wavelet in the wavelet analysis of various times series made by combining white noise and/or sine function. The result derived is also applied to short-memory arctic oscillation index (AOI) and long-memory southern oscillation index (SOI). This study, different from previous studies evaluating one or two mother wavelets, considers a total of four generally-used mother wavelets, Bump, Morlet, Paul, and Mexican Hat. Summarizing the results is as follows. First, the Bump mother wavelet is found to have some limitations to represent the unstationary behavior of the periodic components. Its application results are more or less the same as the spectrum analysis. On the other hand, the Morlet and Paul mother wavelets are found to represent the non-stationary behavior of the periodic components. Finally, the Mexican Hat mother wavelet is found to be too complicated to interpret. Additionally, it is also found that the application result of Paul mother wavelet can be inconsistent for some specific time series. As a result, the Morlet mother wavelet seems to be the most stable one for general applications, which is also assured by the recent trend that the Morlet mother wavelet is most frequently used in the wavelet analysis research.

• The Korean Journal of Food And Nutrition
• /
• v.29 no.3
• /
• pp.382-391
• /
• 2016

The purpose of this experiment was designed to investigate the effects of medicinal herbs (MH) extracts on dementia induced by trimethyltin chloride (TMT) in rats. Six-week-old male Sprague-Dawley rats were randomly divided into five groups; normal group (group 1), control group (group 2), MH extracts group (250, 500 mg/kg) (group 3, group 4) and positive control group (tacrine group, group 5). In the control group to induce dementia, a 2.5 mg/kg of TMT intraperitoneal injection was used for 14 days (1 per day) in the rats. In the MH extracts group 250 mg/kg and 500 mg/kg of MH extracts were medicated in an oral inoculation for 20 days (1 per day). After 30 minutes, a 2.5 mg/kg of TMT intraperitoneal injection, which causes dementia, was used for 14 days (1 per day). In the positive control group (Tacrine group) 10 mg/kg of Tacrine, the dementia treatment, was medicated in an oral inoculation. After 30 mintues, 1 mg/kg of TMT intraperitoneal injection, which causes dementia, was used for 14 days (1 per day). The present author observed the passive avoidance performance test, and memory ability test (Y maze test), the values of MDA, acetlycholinesterase (AchE) activity in the brain and antioxidant enzyme in serum. MH extracts significantly improved memory of AD model rats in the Y-maze test, and also significantly improved memory of AD model rats in the passive avoidance test. MH extracts significantly reduced AChE activity, and significantly increased the SOD level, but not catalase and MDA. From the results above, MH extracts is thought to be effective in the improvement of antioxidant enzymes and memory ability.

• Management & Information Systems Review
• /
• v.33 no.1
• /
• pp.139-155
• /
• 2014

This study is to examine the linkage of volatility between changes in the stock market of India and other countries through the integration of the world economy. The results were as follows: First, autocorrelation or serial correlation did not exist in the classic RS model, but long-term memory was present in the modified RS model. Second, unit root did not exist in the unit root test for all periods, and the series were a stable explanatory power and a long-term memory with the normal conditions in the ARFIMA model. Third, in the multivariate asymmetric BEKK and VAR model before the financial crisis, it showed that there was a strong influence of the own market of Taiwan and UK in the conditional mean equation, and a strong spillover effect from Japan to India, from Taiwan to China(Korea, US), from US(Japan) to UK in one direction. In the conditional variance equation, GARCH showed a strong spillover effect that indicated the same direction as the result of ARCH coefficient of the market itself. Asymmetric effects in three home markets and between markets existed. Fourth, after the financial crisis, in the conditional mean equation, only the domestic market in Taiwan showed strong influences, and strong spillover effects existed from India to US, from Taiwan to Japan, from Korea to Germany in one direction. In the conditional variance equation, strong spillover effects were the same as the result of the pre-crisis and asymmetric effect in the domestic market in UK was present, and one-way asymmetric effect existed in Germany from Taiwan. Therefore, the results of this study presented the linkage between the volatilities of the stock market of India and other countries through the integration of the world economy, observing and confirming the asymmetric reactions and return(volatility) spillover effects between the stock market of India and other countries.