• The Korean Journal of Financial Management
• /
• v.25 no.4
• /
• pp.79-116
• /
• 2008

This paper examines and estimates GARCH-VaR models (RiskMetrics, GARCH, IGARCH, GJR and APARCH) with three different distributions such as Gaussian normal, Student-t, Skewness Student-t Distribution using the daily price data from Korean Stock Market during Jan. 1, 1980-Sept. 30, 2004. It also compares them. In-sample test, this finds that for all confidence level as $90%{\sim}99.9%$, the performance and accuracy of IGARCH with ${\lambda}=0.87$ and skewness Student-t distribution are superior to other models and distributions in long position, but GARCH and GJR with Skewness Student-t distribution in short position. For above 99% confidence level, the performance and accuracy of IGARCH with ${\lambda}=0.87$ in both long and short positions are superior to other models and distributions, but Skewness Student-t distribution for long position and Student-t distribution for short position are more accuracy and superior to other distributions. In-out-of sample test, these results also confirm the evidences that the above findings are consistent as well.

• The Korean Journal of Applied Statistics
• /
• v.29 no.4
• /
• pp.753-771
• /
• 2016

Value at Risk (VaR) is widely used as an important tool for risk management of financial institutions. In this paper we discuss estimation and back testing for VaR of the portfolio composed of KOSPI, Dow Jones, Shanghai, Nikkei indexes. The copula functions are adopted to construct the multivariate distributions of portfolio components from marginal distributions that combine extreme value theory and GARCH models. Volatility models with t distribution of the error terms using Gaussian, t, Clayton and Frank copula functions are shown to be more appropriate than the other models, in particular the model using the Frank copula is shown to be the best.

• The Korean Journal of Applied Statistics
• /
• v.25 no.1
• /
• pp.29-43
• /
• 2012

Financial data such as stock index returns and exchange rates have the properties of heavy tail and asymmetry compared to normal distribution. When we estimate VaR using the GARCH model (with the conditional return distribution of normal) it shows the tendency of the lower estimation and clustering in the losses over the estimated VaR. In this paper, we argue that this problem can be resolved through the adaptation of the unbounded Johnson distribution as that of the condition return. We also compare this model with the GARCH with the conditional return distribution of normal and student-t. Using the losses exceed the ex-ante VaR, estimates, we check the validity of the GARCH models through the failure proportion test and the clustering test. We nd that the GARCH model with conditional return distribution of unbounded Johnson provides an appropriate estimation of the VaR and does not occur the clustering of violations.

• The Korean Journal of Applied Statistics
• /
• v.24 no.1
• /
• pp.61-69
• /
• 2011

In GARCH context, the conditional variance (or volatility) is of a quadratic function of the observation process. Examine standard ARCH/GARCH and their variant models in terms of quadratic formulations and it is interesting to note that most models in GARCH context have contained neither the first order term nor the interaction term. In this paper, we consider three models possessing the first order and/or interaction terms in the formulation of conditional variances, viz., quadratic GARCH, absolute value GARCH and bilinear GARCH processes. These models are investigated with a view to model comparisons and applications to financial time series in Korea

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.6
• /
• pp.1041-1049
• /
• 2010

In Korean professional baseball, attendance is the largest source of revenue for development of professional baseball and the highest concern of professional baseball teams. So, if there is demand forecasting model, it will be helpful for pennant chasers to work out the strategies for drawing attendance. For this reason, this research intends to suggest the model which estimates Korean professional baseball's attendance and uses all usable variables which have an effect on attendance in limited circumstances. We supposed that dependent variable is attendance as well as several independent variables and error term are homoscedastic variance. And then, we compared the models which assume conditional heteroscedastic variance like GARCH and EGARCH with GARCH-t models which use the assumption that error term's distribution follows student-t distribution. In result of that, we could confirm that the models which were made by using GARCH(1,1)-t made estimates the most accurately among the several models considered.

• The Korean Journal of Applied Statistics
• /
• v.23 no.2
• /
• pp.219-234
• /
• 2010

In index investing according to KOSPI, we estimate Value at Risk(VaR) from the extreme losses of the daily returns which are obtained from KOSPI. To this end, we apply Block Maxima(BM) model which is one of the useful models in the extreme value theory. We also estimate the extremal index to consider the dependency in the occurrence of extreme losses. From the back-testing based on the failure rate method, we can see that the model is adaptable for the VaR estimation. We also compare this model with the GARCH model which is commonly used for the VaR estimation. Back-testing says that there is no meaningful difference between the two models if we assume that the conditional returns follow the t-distribution. However, the estimated VaR based on GARCH model is sensitive to the extreme losses occurred near the epoch of estimation, while that on BM model is not. Thus, estimating the VaR based on GARCH model is preferred for the short-term prediction. However, for the long-term prediction, BM model is better.

• Environmental and Resource Economics Review
• /
• v.16 no.4
• /
• pp.947-978
• /
• 2007

In this paper, we investigated a Value-at-Risk approach to the volatility of two crude oil markets (Brent and Dubai). We also assessed the performance of various VaR models (RiskMetrics, GARCH, IGARCH and FIGARCH models) with the normal and skewed Student-t distribution innovations. The FIGARCH model outperforms the GARCH and IGARCH models in capturing the long memory property in the volatility of crude oil markets returns. This implies that the long memory property is prevalent in the volatility of crude oil returns. In addition, from the results of VaR analysis, the FIGARCH model with the skewed Student-t distribution innovation predicts critical loss more accurately than other models with the normal distribution innovation for both long and short positions. This finding indicates that the skewed Student-t distribution innovation is better for modeling the skewness and excess kurtosis in the distribution of crude oil returns. Overall, these findings might improve the measurement of the dynamics of crude oil prices and provide an accurate estimation of VaR for buyers and sellers in crude oil markets.

• 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 Korea Port Economic Association
• /
• v.31 no.1
• /
• pp.1-14
• /
• 2015

The standard GARCH model imposing symmetry on the conditional variance, tends to fail in capturing some important features of the data. This paper, hence, introduces the models capturing asymmetric effect. They are the EGARCH model and the GJR model. We provide the systematic comparison of volatility models focusing on the asymmetric effect of news on volatility. Specifically, three diagnostic tests are provided: the sign bias test, the negative size bias test, and the positive size bias test. This paper shows that there is significant evidence of GARCH-type process in the data, as shown by the test for the Ljung-Box Q statistic on the squared residual data. The estimated unconditional density function for squared residual is clearly skewed to the left and markedly leptokurtic when compared with the standard normal distribution. The observation of volatility clustering is also clearly reinforced by the plot of the squared value of residuals of export volume and values. The unconditional variance of both export volumes and export value indicates that large shocks of either sign tend to be followed by large shocks, and small shocks of either sign tend to follow small shocks. The estimated export volume news impact curve for the GARCH also suggests that $h_t$ is overestimated for large negative and positive shocks. The conditional variance equation of the GARCH model for export volumes contains two parameters ${\alpha}$ and ${\beta}$ that are insignificant, indicating that the GARCH model is a poor characterization of the conditional variance of export volumes. The conditional variance equation of the EGARCH model for export value, however, shows a positive sign of parameter ${\delta}$, which is contrary to our expectation, while the GJR model exhibits that parameters ${\alpha}$ and ${\beta}$ are insignificant, and ${\delta}$ is marginally significant. That indicates that the asymmetric volatility models are poor characterization of the conditional variance of export value. It is concluded that the asymmetric EGARCH and GJR model are appropriate in explaining the volatility of export volume, while the symmetric standard GARCH model is good for capturing the volatility.

• The Korean Journal of Applied Statistics
• /
• v.33 no.6
• /
• pp.713-722
• /
• 2020

Contrasted with the standard symmetric GARCH models, we consider a broad class of threshold-asymmetric models to analyse financial time series exhibiting asymmetric volatility. By further introducing power transformations, we add more flexibilities to the asymmetric class, thereby leading to power transformed and asymmetric volatility models. In particular, the paper is concerned with the nonstationary volatilities in which conditions for integrated volatility and explosive volatility are separately discussed. Dow Jones Industrial Average is analysed for illustration.