• Journal of the Korea Academia-Industrial cooperation Society
• /
• 제13권9호
• /
• pp.3900-3914
• /
• 2012

This study examined the background of the recent global financial crisis and the concept of one of the financial derivatives such as the credit default swap(CDS) or synthetic CDO(collateral debt obligations), given the rapid growing and changing the over-the-counter derivative markets in their volume and structures. In comparison with the previous literature such as the study of Park & Kim (2011), this research empirically performed more thorough and comprehensive investigations to find any financial characteristics or attributes to determine the CDS spreads. Regarding the results obtained from the multiple regression models, the explanatory variables such as STYIELD3, SLOPE, INASSETS, and VOLATILITY, showed their statistically significant effects on all the tested dependent variables(DVs). Another procedure such as the principle component analysis(PCA), was also performed to account for additional IDVs as possible determinants of the dependent variables. Subsequent to this analysis, larger coefficients of each corresponding eigenvector such as BETA, PFT2, GROWTH, STD, and BLEVERAGE were found to be possible financial determinants. For robustness, all the IDVs were employed to be tested in the 'full' regression model with stepwise procedure. As a result, STYIELD3, SLOPE, and VOLATILITY, and BETA showed their statistically significant relationship with all the dependent variables of the CDS spreads.

• Journal of Intelligence and Information Systems
• /
• 제29권1호
• /
• pp.377-396
• /
• 2023

As the cryptocurrency market continues to grow, it has developed into a new financial market. The need for investment strategy research on the cryptocurrency market is also emerging. This study aims to conduct an empirical analysis on an investment methodology of cryptocurrency that combines short-term trading strategy and deep learning. Daily price data of the Ethereum was collected through the API of Upbit, the Korean cryptocurrency exchange. The investment performance of the experimental model was analyzed by finding the optimal parameters based on past data. The experimental model is a volatility breakout strategy(VBS), a Long Short Term Memory(LSTM) model, moving average cross strategy and a combined model. VBS is a short-term trading strategy that buys when volatility rises significantly on a daily basis and sells at the closing price of the day. LSTM is suitable for time series data among deep learning models, and the predicted closing price obtained through the prediction model was applied to the simple trading rule. The moving average cross strategy determines whether to buy or sell when the moving average crosses. The combined model is a trading rule made by using derived variables of the VBS and LSTM model using AND/OR for the buy conditions. The result shows that combined model is better investment performance than the single model. This study has academic significance in that it goes beyond simple deep learning-based cryptocurrency price prediction and improves investment performance by combining deep learning and short-term trading strategies, and has practical significance in that it shows the applicability in actual investment.

• The Journal of Bigdata
• /
• 제8권1호
• /
• pp.23-34
• /
• 2023

Nowadays, the Designated Driver Services employ dynamic pricing, which adapts in real-time based on nearby driver availability, service user volume, and current weather conditions during the user's request. The uncertain volatility is the main cause of price increases, leading to customer attrition and service refusal from driver. To make a good Designated Driver Services, development of a demand forecasting model is required. In this study, we propose developing a demand forecasting model using data from the Designated Driver Service by considering normal and peak periods, such as rush hour and rush day, as prior knowledge to enhance the model performance. We propose a new methodology called Time-Series with Conditional Probability(TSCP), which combines conditional probability and time-series models to enhance performance. Extensive experiments have been conducted with real Designated Driver Service data, and the result demonstrated that our method outperforms the existing time-series models such as SARIMA, Prophet. Therefore, our study can be considered for decision-making to facilitate proactive response in Designated Driver Services.

• Proceedings of the Korea Air Pollution Research Association Conference
• /
• 한국대기환경학회 2006년도 춘계학술대회 논문집
• /
• pp.55-56
• /
• 2006

This thesis describes the use of measured aerosol size distributions and size-resolved hygroscopic growth to examine the physical and chemical properties of several particle classes. The primary objective of this work was to investigate the optical and cloud forming properties of a range of ambient aerosol types measured in a number of different locations. The tool used for most of these analyses is a differential mobility analyzer / tandem differential mobility analyzer (DMA / TDMA) system developed in our research group. To collect the data described in two of the chapters of this thesis, an aircraft-based version of the DMA / TDMA was deployed to Japan and California. The data described in two other chapters were conveniently collected during a period when the aerosol of interest came to us. The unique aspect of this analysis is the use of these data to isolate the size distributions of distinct aerosol types in order to quantify their optical and cloud forming properties. I used collected data during the Asian Aerosol Characterization Experiment (ACE-Asia) to examine the composition and homogeneity of a complex aerosol generated in the deserts and urban regions of China and other Asian countries. An aircraft-based tandem differential mobility analyzer was used for the first time during this campaign to examine the size-resolved hygroscopic properties of particles having diameters between 40 and 586 nm. Asian Dust Above Monterey (ADAM-2003) study was designed both to evaluate the degree to which models can predict the long-range transport of Asian dust, and to examine the physical and optical properties of that aged dust upon reaching the California coast. Aerosol size distributions and hygroscopic growth are measured in College Station, TX to investigate the cloud nucleating and optical properties of a biomass burning aerosol generated from fires on the Yucatan Peninsula. Measured aerosol size distributions and size-resolved hygroscopicity and volatility were used to infer critical supersaturation distributions of the distinct particle types that were observed during this period. The predicted CCN concentrations were used in a cloud model to determine the impact of the different aerosol types on the expected cloud droplet concentration. RH-dependent aerosol extinction coefficients are calculated at a wavelength of 550 nm.

• Journal of Distribution Science
• /
• 제12권3호
• /
• pp.65-73
• /
• 2014

Purpose - Aging and fewer economically active people have challenged the assumption of continuous population increases. A new real estate valuation methodology reflecting changes in population structure is thus needed. Research design, data, and methodology - The relationship between demographic change and changes in real estate prices is analyzed using ordinary least squares (OLS) to estimate the parameters, and a population structure change (PSC)-Binomial Option Model is developed to assess the volatility of the estimated parameters. Results based on Seoul and Shanghai data are compared. Results - Results of the DCF method indicate that investing in Seoul is better than investing in Shanghai, but the binomial option indicates the opposite. The PSC-binomial option model, reflecting changes in population structure, yields higher values (24.6 million won in Seoul and 43.3 million won in Shanghai) than those given by the binomial option model. Conclusions - This study indicates that applying changes in population structure to existing research, such as in the binomial option model, represents a more accurate real estate valuation method. Results demonstrate that the new model is more accurate than existing models such as the DCF or binomial option.

• The Korean Journal of Financial Management
• /
• 제22권1호
• /
• pp.93-118
• /
• 2005

This paper examines credit spreads in Korea corporate market using one of structural models, the mean reverting leverage ratio model (Collin-Dufresne and Goldstein (2001)). Compared to the actual credit spreads, we show that the credit spreads induced by the model are overpredicted. We also investigate the systematic errors that cause the over-pre-diction of credit spreads using the t-test. We show that the systematic errors are affected by the current leverage ratio and asset volatility.

• The Korean Journal of Financial Management
• /
• 제26권3호
• /
• pp.227-246
• /
• 2009

This study examines the dynamic hedging performances of the Black-Scholes model and Heston model when stock prices drift with stochastic volatilities. Using Monte Carlo simulations, stock prices consistent with Heston's(1993) stochastic volatility option pricing model are generated. In this circumstance, option traders are assumed to use the Black- Scholes model and Heston model to implement dynamic hedging strategies for the options written. The results of simulation indicate that the hedging performance of a mis-specified Black-Scholes model is almost as good as that of a fully specified Heston model. The implication of these results is that the efficacy of the dynamic hedging performances on evaluating the specifications of alternative option models can be limited.

• Journal of the Society of Naval Architects of Korea
• /
• 제58권5호
• /
• pp.281-293
• /
• 2021

Oil and steel prices, which are major pricescosts in the shipbuilding industry, were predicted. Firstly, the error of the moving average line (N=3-5) was examined, and in all three error analyses, the moving average line (N=3) was small. Secondly, in the linear prediction of data through existing theory, oil prices rise slightly, and steel prices rise sharply, but in reality, linear prediction using existing data was not satisfactory. Thirdly, we identified the limitations of linear prediction methods and confirmed that oil and steel price prediction was somewhat similar to actual moving average line prediction methods. Due to the high volatility of major price flows, large errors were inevitable in the forecast section. Through the time series analysis method at the end of this paper, we were able to achieve not bad results in all analysis items relative to artificial intelligence (Prophet). Predictive data through predictive analysis using eight predictive models are expected to serve as a good research foundation for developing unique tools or establishing evaluation systems in the future. This study compares the basic settings of artificial intelligence programs with the results of core price prediction in the shipbuilding industry through time series prediction theory, and further studies the various hyper-parameters and event effects of Prophet in the future, leaving room for improvement of predictability.

• KIPS Transactions on Software and Data Engineering
• /
• 제11권1호
• /
• pp.19-28
• /
• 2022

Recently, cryptocurrency has attracted much attention, and price prediction studies of cryptocurrency have been actively conducted. Especially, efforts to improve the prediction performance by applying the deep learning model are continuing. LSTM (Long Short-Term Memory) model, which shows high performance in time series data among deep learning models, is applied in various views. However, it shows low performance in cryptocurrency price data with high volatility. Although, to solve this problem, new input features were found and study was conducted using them, there is a lack of study on input features that drop predictive performance. Thus, in this paper, we collect the recent trends of six cryptocurrencies including Bitcoin and Ethereum and analyze effects of input features on the cryptocurrency price predictive performance through statistics and deep learning. The results of the experiment showed that cryptocurrency price predictive performance the best when open price, high price, low price, volume and price were combined except for rate of closing price fluctuation.

• The Korean Journal of Applied Statistics
• /
• 제37권1호
• /
• pp.39-48
• /
• 2024

In this paper, we introduce an extended GARCH model designed to address asymmetric leverage effects. The variance in the standard GARCH model is composed of past conditional variances and past squared residuals. However, it is not possible to model asymmetric leverage effects with squared residuals alone, so in this paper, we propose a new extended GARCH model to explain the leverage effects using the Yeo-Johnson transformation which adjusts transformation parameter to make asymmetric data more normal or symmetric. We utilize the reverse properties of Yeo-Johnson transformation to model asymmetric volatility. We investigate the characteristics of the proposed model and parameter estimation. We also explore how to derive forecasts and forecast intervals in the proposed model. We compare it with standard GARCH and other extended GARCH models that model asymmetric leverage effects through empirical data analysis.