• Journal of environmental and Sanitary engineering
• /
• v.23 no.4
• /
• pp.83-93
• /
• 2008

The non-linear least squares model(NLSM) has long been the standard technique used by hydrologists for constructing rating curves. The reasons for its adaptation are vague, and its appropriateness as a method of describing discharge measurement uncertainty has not been well investigated. It is shown in this paper that the classical method of NLSM can model only a very limited class of variance heterogeneity. Furthermore, this lack of flexibility often leads to unaccounted heteroscedasticity, resulting in dubious values for the rating curve parameters and estimated discharge. By introducing a heteroscedastic maximum likelihood model(HMLM), the variance heterogeneity is treated more generally. The maximum likelihood model stabilises the variance better than the NLSM approach, and thus is a more robust and appropriate way to fit a rating curve to a set of discharge measurements.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.1
• /
• pp.329-338
• /
• 2019

In this study, we analyze consumer preferences based on the agricultural box scheme attributes, and make a suggestion for business revival. We estimate the marginal willingness to pay (MWTP) for box scheme attributes using a choice experiment. Attributes include the bundle method, the delivery method, and price. To select an efficient model for statistical analysis, we evaluate the conditional logit model, heteroscedastic extreme value model(HEV model), multinomial probit model, and mixed logit model under different assumptions. The results of these four models show that the bundle method, the delivery method, and price are statistically significant in explaining the probability of participation in a box scheme. The results of likelihood ratio tests show that the heteroscedastic extreme value model is the most appropriate for our survey data. The results also indicate that MWTP for a change from fixed type to selection type is KRW 7,096.6. MWTP for a change from parcel service to direct delivery and cold-chain delivery are KRW 3,497.5 and KRW 7,532.7, respectively. The results of this study may contribute to the government's local food policies.

• The Korean Journal of Applied Statistics
• /
• v.15 no.2
• /
• pp.311-321
• /
• 2002

Transfer function model(TFM) capturings conditional heteroscedastic pattern is introduced to analyze stochastic regression relationship between the two time series. Nonlinear ARCH concept is incorporated into the TFM via threshold ARCH and beta- ARCH models. Steps for statistical analysis of the proposed model are explained along the lines of the Box & Jenkins(1976, ch. 10). For illustration, dynamic analysis between KOSPI and NASDAQ is conducted from which it is seen that threshold ARCH performs the best.

• The Korean Journal of Applied Statistics
• /
• v.24 no.6
• /
• pp.1033-1043
• /
• 2011

In this paper, we use a nested family of models of Generalized Autoregressive Conditional Heteroscedasticity(GARCH) to verify asymmetric conditional heteroscedasticity in the KOSPI and Won-Dollar exchange rate. This study starts from an investigation of whether time series data have asymmetric features not explained by standard GARCH models. First, we use kernel density plot to show the non-normality and asymmetry in data as well as to capture asymmetric conditional heteroscedasticity. Later, we use three representative asymmetric heteroscedastic models, EGARCH(Exponential Garch), GJR-GARCH(Glosten, Jagannathan and Runkle), APARCH(Asymmetric Power Arch) that are improved from standard GARCH models to give a better explanation of asymmetry. Thereby we highlight the fact that volatility tends to respond asymmetrically according to positive and/or negative values of past changes referred to as the leverage effect. Furthermore, it is verified that how the direction of asymmetry is different depending on characteristics of time series data. For the KOSPI and Korean won-US dollar exchange rate, asymmetric heteroscedastic model analysis successfully reveal the leverage effect. We obtained predictive values of conditional volatility and its prediction standard errors by using moving block bootstrap.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.44 no.7
• /
• pp.1-13
• /
• 2002

Two global optimization methods, the SCE-UA method and the Annealing-simplex (A-S) method for calibrating a daily rainfall-runoff model, a Tank model, was compared with that of the Downhill Simplex method. The performance of the four objective functions, DRMS (daily root mean square), HMLE (heteroscedastic maximum likelihood estimator), ABSERR (mean absolute error), and NS (Nash-Sutcliffe measure), was tested and synthetic data and historical data were used. In synthetic data study. 100% success rates for all objective functions were obtained from the A-S method, and the SCE-UA method was also consistently able to obtain good estimates. The downhill simplex method was unable to escape from local optimum, the worst among the methods, and converged to the true values only when the initial guess was close to the true values. In the historical data study, the A-S method and the SCE-UA method showed consistently good results regardless of objective function. An objective function was developed with combination of DRMS and NS, which putted more weight on the low flows.

• MALSORI
• /
• no.44
• /
• pp.105-113
• /
• 2002

In Korean, each digit is monosyllable and some pairs are known to have high confusability, causing performance degradation of connected digit recognition systems. To improve the performance, in this paper, we employ various discriminant analyses (DA) including Linear DA (LDA), Weighted Pairwise Scatter LDA WPS-LDA), Heteroscedastic Discriminant Analysis (HDA), and Maximum Likelihood Linear Transformation (MLLT). We also examine several combinations of various DA for additional performance improvement. Experimental results show that applying any DA mentioned above improves the string accuracy, but the amount of improvement of each DA method varies according to the model complexity or number of mixtures per state. Especially, more than 20% of string error reduction is achieved by applying MLLT after WPS-LDA, compared with the baseline system, when class level of DA is defined as a tied state and 1 mixture per state is used.

• Communications for Statistical Applications and Methods
• /
• v.17 no.2
• /
• pp.255-261
• /
• 2010

A class of threshold-asymmetric GRACH(TGARCH, hereafter) models has been useful for explaining asymmetric volatilities in the field of financial time series. The cumulative impulse response function of a conditionally heteroscedastic time series often measures a degree of unstability in volatilities. In this article, a general form of the cumulative impulse response function of the TGARCH model is discussed. In particular, We present formula in their closed forms for the first two lower order models, viz., TGARCH(1, 1) and TGARCH(2, 2).

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.3
• /
• pp.547-557
• /
• 2017

The health-related quality of life data are commonly skewed and bounded with spike at the perfect health status, and the variance tended to be heteroscedastic. In this study, we have developed a prediction model for EQ-5D using linear regression model, beta regression model, and extended beta regression model with mean and precision submodel, and also compared the predictive accuracy. The extended beta regression model allows to model skewness and differences in dispersion related to covariates. Although the extended beta regression model has higher prediction accuracy than the linear regression model, the overlapped confidence intervals suggested that the extended beta regression model was superior to the linear regression model. However, the expended beta regression model could explain the heteroscedasticity and predict within the bounded range. Therefore, the expended beta regression model are appropriate for fitting the health-related quality of life data such as EQ-5D.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.1
• /
• pp.89-96
• /
• 2009

We examined the characteristics of temperature variation in Seoul between 1961 to 2008 using the spectral heteroscedastic model. The mean function in the propsed model explains the season effect using periodic functions and the overall increase using the quadratic regression spline. The variance function also had periodic functions to explain the seasonality of variance. We found that there has been annual mean temperature increase by about $1.5^{\circ}C$ for the last 48 years. The increase of annual mean temperature was mainly caused by the increase in winter, which made the amplitude decreased.

• Communications for Statistical Applications and Methods
• /
• v.16 no.2
• /
• pp.287-297
• /
• 2009

In terms of generalized autoregressive conditional heteroscedastic(GARCH) model, estimation of prediction interval based on likelihood is quite sensitive to distribution of error. Moveover, it is not an easy job to construct prediction interval for conditional variance. Recent studies show that the bootstrap method can be one of the alternatives for solving the problems. In this paper, we introduced the bootstrap approach proposed by Pascual et al. (2006). We employed it to Korean stock price data set.