• Journal of the Korean Data and Information Science Society
• /
• v.28 no.4
• /
• pp.797-810
• /
• 2017

The Korean peninsula is exposed to typhoons every year. Typhoons cause huge socioeconomic damage because tropical cyclones tend to occur with strong winds and heavy precipitation. In order to understand the complex dependence structure between strong winds and heavy precipitation, the copula links a set of univariate distributions to a multivariate distribution and has been actively studied in the field of hydrology. In this study, we carried out analysis using data of wind speed and precipitation collected from the weather stations in Busan and Jeju. Log-Normal, Gamma, and Weibull distributions were considered to explain marginal distributions of the copula. Kolmogorov-Smirnov, Cramer-von-Mises, and Anderson-Darling test statistics were employed for testing the goodness-of-fit of marginal distribution. Observed pseudo data were calculated through inverse transformation method for establishing the copula. Elliptical, archimedean, and extreme copula were considered to explain the dependence structure between strong winds and heavy precipitation. In selecting the best copula, we employed the Cramer-von-Mises test and cross-validation. In Busan, precipitation according to average wind speed followed t copula and precipitation just as maximum wind speed adopted Clayton copula. In Jeju, precipitation according to maximum wind speed complied Normal copula and average wind speed as stated in precipitation followed Frank copula and maximum wind speed according to precipitation observed Husler-Reiss copula.

• Journal of the Korea Society for Simulation
• /
• v.28 no.2
• /
• pp.119-129
• /
• 2019

Although line production systems with finite buffers have been studied over several decades, except for some special cases there are no explicit expressions for system performances such as waiting times(or response time) and blocking probability. Recently, a max-plus algebraic approach for buffer-sharing systems with constant processing times was introduced and it can lead to analytic expressions for (higher) moment and tail probability of stationary waiting. Theoretically this approach can be applied to general processing times, but it cannot give a proper way for computing performance measures. To this end, in this study we developed simulation models using @RISK software and the expressions derived from max-plus algebra, and computed and compared blocking probability, waiting time (or response time) with respect to two blocking policies: communication(BBS: Blocking Before Service) and production(BAS: Blocking After Service). Moreover, an optimization problem which determines the minimum shared-buffer capacity satisfying a predetermined QoS(quality of service) is also considered.

• The Korean Journal of Financial Management
• /
• v.26 no.3
• /
• pp.139-169
• /
• 2009

This paper introduces the modified VaR which takes into account the asymmetry and fat-tails of financial asset distribution, and then compares its out-of-sample forecast performance with traditional VaR model such as historical simulation model and Riskmetrics. The empirical tests using stock indices of 6 countries showed that the modified VaR has the best forecast accuracy. At the test of independence, Riskmetrics and GARCH model showed best performances, but the independence was not rejected for the modified VaR. The Monte Carlo simulation using skew t distribution again proved the best forecast performance of the modified VaR. One of many advantages of the modified VaR is that it is appropriate for measuring VaR of the portfolio, because it can reflect not only the linear relationship but also the nonlinear relationship between individual assets of the portfolio through coskewness and cokurtosis. The empirical analysis about decomposing VaR of the portfolio of 6 stock indices confirmed that the component VaR is very useful for the re-allocation of component assets to achieve higher Sharpe ratio and the active risk management.

• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.107-122
• /
• 2017

Volatility in the stock market returns is a measure of investment risk. It plays a central role in portfolio optimization, asset pricing and risk management as well as most theoretical financial models. Engle(1982) presented a pioneering paper on the stock market volatility that explains the time-variant characteristics embedded in the stock market return volatility. His model, Autoregressive Conditional Heteroscedasticity (ARCH), was generalized by Bollerslev(1986) as GARCH models. Empirical studies have shown that GARCH models describes well the fat-tailed return distributions and volatility clustering phenomenon appearing in stock prices. The parameters of the GARCH models are generally estimated by the maximum likelihood estimation (MLE) based on the standard normal density. But, since 1987 Black Monday, the stock market prices have become very complex and shown a lot of noisy terms. Recent studies start to apply artificial intelligent approach in estimating the GARCH parameters as a substitute for the MLE. The paper presents SVR-based GARCH process and compares with MLE-based GARCH process to estimate the parameters of GARCH models which are known to well forecast stock market volatility. Kernel functions used in SVR estimation process are linear, polynomial and radial. We analyzed the suggested models with KOSPI 200 Index. This index is constituted by 200 blue chip stocks listed in the Korea Exchange. We sampled KOSPI 200 daily closing values from 2010 to 2015. Sample observations are 1487 days. We used 1187 days to train the suggested GARCH models and the remaining 300 days were used as testing data. First, symmetric and asymmetric GARCH models are estimated by MLE. We forecasted KOSPI 200 Index return volatility and the statistical metric MSE shows better results for the asymmetric GARCH models such as E-GARCH or GJR-GARCH. This is consistent with the documented non-normal return distribution characteristics with fat-tail and leptokurtosis. Compared with MLE estimation process, SVR-based GARCH models outperform the MLE methodology in KOSPI 200 Index return volatility forecasting. Polynomial kernel function shows exceptionally lower forecasting accuracy. We suggested Intelligent Volatility Trading System (IVTS) that utilizes the forecasted volatility results. IVTS entry rules are as follows. If forecasted tomorrow volatility will increase then buy volatility today. If forecasted tomorrow volatility will decrease then sell volatility today. If forecasted volatility direction does not change we hold the existing buy or sell positions. IVTS is assumed to buy and sell historical volatility values. This is somewhat unreal because we cannot trade historical volatility values themselves. But our simulation results are meaningful since the Korea Exchange introduced volatility futures contract that traders can trade since November 2014. The trading systems with SVR-based GARCH models show higher returns than MLE-based GARCH in the testing period. And trading profitable percentages of MLE-based GARCH IVTS models range from 47.5% to 50.0%, trading profitable percentages of SVR-based GARCH IVTS models range from 51.8% to 59.7%. MLE-based symmetric S-GARCH shows +150.2% return and SVR-based symmetric S-GARCH shows +526.4% return. MLE-based asymmetric E-GARCH shows -72% return and SVR-based asymmetric E-GARCH shows +245.6% return. MLE-based asymmetric GJR-GARCH shows -98.7% return and SVR-based asymmetric GJR-GARCH shows +126.3% return. Linear kernel function shows higher trading returns than radial kernel function. Best performance of SVR-based IVTS is +526.4% and that of MLE-based IVTS is +150.2%. SVR-based GARCH IVTS shows higher trading frequency. This study has some limitations. Our models are solely based on SVR. Other artificial intelligence models are needed to search for better performance. We do not consider costs incurred in the trading process including brokerage commissions and slippage costs. IVTS trading performance is unreal since we use historical volatility values as trading objects. The exact forecasting of stock market volatility is essential in the real trading as well as asset pricing models. Further studies on other machine learning-based GARCH models can give better information for the stock market investors.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.35 no.10
• /
• pp.899-904
• /
• 2007

During the conceptual design phase of fighter class aircraft, the high speed wind tunnel test with 1/20 scale wing-body-tail model has been conducted to investigate the effects of leading edge flap deflection on the supersonic cruise performance of the aircraft. To select the proper leading edge flap deflection for the wind tunnel test, the aerodynamic characteristics due to various leading edge flap deflections have been analyzed by using corrected supersonic panel method. Based on the results obtained from the experimental and numerical approaches, the effects of leading edge flap deflection have shown to be useful to enhance the supersonic cruise performance of fighter class aircraft.

• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.793-808
• /
• 2017

We compare various variable screening methods on multiclass classification problems when the data is ultrahigh-dimensional. Two different approaches were considered: (1) pairwise extension from binary classification via one versus one or one versus rest comparisons and (2) direct classification of multiclass responses. We conducted extensive simulation studies under different conditions: heavy tailed explanatory variables, correlated signal and noise variables, correlated joint distributions but uncorrelated marginals, and unbalanced response variables. We then analyzed real data to examine the performance of the methods. The results showed that model-free methods perform better for multiclass classification problems as well as binary ones.

• The Korean Journal of Applied Statistics
• /
• v.31 no.1
• /
• pp.109-122
• /
• 2018

The composite lognormal-generalized Pareto distribution (LN-GPD) is a mixture of right-truncated lognormal and GPD for a given threshold value. Scollnik (Scandinavian Actuarial Journal, 2007, 20-33, 2007) shows that the composite LN-GPD is adequate to describe body distribution and heavy-tailedness. This paper considers time-varying modeling of the LN-GPD based on local polynomial maximum likelihood estimation. Time-varying model provides significant detailed information of time dependent data, hence it can be applied to disciplines such as service engineering for staffing and resources management. Our work also extends to Beirlant and Goegebeur (Journal of Multivariate Analysis, 89, 97-118, 2004) in the sense of losing no data by including truncated lognormal distribution. Our proposed method is shown to perform adequately in simulation. Real data application to the service time of the Israel bank call center shows interesting findings on the staffing policy.

• The Korean Journal of Applied Statistics
• /
• v.36 no.6
• /
• pp.591-618
• /
• 2023

This paper evaluates the performance of various distribution assumptions in regression settings for estimating insurance loss. The gamma distribution is commonly used to handle the asymmetry property of loss distribution. However, recent studies highlight the significance of heavy-tailedness in loss distribution. Through an analysis of real home fire insurance data, we compare the effectiveness of different distribution assumptions in regression methods. Our findings show that the choice of parametric distributional assumption is crucial in determining premiums for various insurance products, including "excess of loss insurance" and "limit insurance". Additionally, we discuss practical considerations for applying our results in home fire insurance.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.12 no.3
• /
• pp.125-132
• /
• 2007

We estimated the distribution of predator-prey interaction strengths for 12 species of herbivores (including amphipods, isopods, gastropods, and sea urchins) and made a regression model that may be applicable to other species. Laboratory experiments were used to determine per capita grazing rate (PCGR; g seaweeds/individual/day). Relationship between the biomass of individual grazers and fourth-root transformed PCGR was fitted to power curve ($y=0.2310x^{0.3290}$, r=0.8864). This finding supported that the grazing efficiency was not even as individual grazers increase in size (biomass). Therefore, the biomass-normalized PCGR was estimated and revealed that smaller size herbivores were more effective grazers. Grazing impact considering density of each taxon was calculated. The sea hare Aplysia kurodai had greatest grazing impact on the seaweed bed and the sea urchin Strongylocentrotus nudus and S. intermedius were ranked in descending order of the impact. The amount of seaweed grazed by the amphipod Elasmopus sp. (>4,000 $ind./m^2$) and Jassa falcata (>2,000 $ind./m^2$) were 3.435 and $1.697mg/m^2/day$ respectively. The combined grazing amount of herbivores was $5,045mg/m^2/day$ in the seaweed bed. Although sea hare and sea urchin had strong impacts on seaweeds, the effects of dense, smaller species could not be seen as negligible. Surprisingly, the calculated grazing potential of sea urchins with a mean density of 3 $ind./m^2$ exceeded the mean production of seaweed cultured in domestic coastal waters in Korea (ca., 5 ton/ha). Small crustaceans were also expected to consume up to 16% of the seaweed production if their densities were rising under weak predation conditions. Considering that the population density of herbivores are strongly controlled by fish, human interference like overfishing may have strong negative effects on persistence of seaweeds communities.

• Korean Journal of Cognitive Science
• /
• v.32 no.1
• /
• pp.1-53
• /
• 2021

The aim of this study was to analyze word frequency effects on eye fixation duration in Korean reading with a one-boundary diffusion model and to show how these phenomena differ between adults (20-28yrs) and adolescents (13-14yrs). We predicted that the drift rate parameter in the boundary diffusion model would reflect the information processing of the fovea during silent reading. Through an eye movement tracking experiment while controlling word properties such as the word frequency and the age of acquisition, Experiment 1 and Experiment 2 show that the information processing pertaining to words to be placed in the fovea is connected to the drift rate of the one-boundary diffusion model parameters. In Experiment 1,in the adult group, the mean difference in the fixation time in the response proportion between the presence of high-frequency condition and low-frequency condition in the adult group was higher in quantile 0.9 than it was in the 0.1 quantile, but in the adolescent group, the mean difference in the fixation time in the response proportion between the two conditions was not significantly in the 0.9 quartile.In Experiment 2, the mean difference in the fixation time in the response proportion between early-acquired condition and late-acquired condition in both groups was also higher in the quantile 0.9 than in the 0.1 quantile. The distribution of the two conditions in the both groups was positively skewed, and the difference showed the same pattern found in the results of Ratcliff(Ratcliff & McKoon, 2008). Based on the experimental results, we propose one-boundary diffusion model as a tool to explain word property effects and individual differences in reading. In particular, we suggest that the drift rate parameter in the boundary diffusion model reflects the information processing of the fovea during reading. In addition, the results show that one-boundary diffusion model can be used to predict the aforementioned phenomena in reading.