• Communications for Statistical Applications and Methods
• /
• v.10 no.3
• /
• pp.873-878
• /
• 2003

LS-SVM(Least-Squares Support Vector Machine) has been used as a promising method for regression as well as classification. Suykens et al.(2000) used only the magnitude of residuals to obtain SVs(Support Vectors). Suykens' method behaves well for homogeneous model. But in a heteroscedastic model, the method shows a poor behavior. The present paper proposes a new method to get SVs. The proposed method uses the variance of noise as well as the magnitude of residuals to obtain support vectors. Through the simulation study we justified excellence of our proposed method.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.5
• /
• pp.905-912
• /
• 2009

We propose a doubly penalized kernel machine (DPKM) which uses heteroscedastic location-scale model as basic model and estimates both mean and volatility functions simultaneously by kernel machines. We also present the model selection method which employs the generalized approximate cross validation techniques for choosing the hyperparameters which affect the performance of DPKM. Artificial examples are provided to indicate the usefulness of DPKM for the mean and volatility functions estimation.

• The Korean Journal of Applied Statistics
• /
• v.6 no.2
• /
• pp.303-309
• /
• 1993

In this paper we have defined two approaches to be error-robust when the precise form of error-structure is unknown. An experiment is optimal by the first criterion if it maximizes the minimum effciency over all candidates of error structure and is optimal by the second if it maximizes the minimum average of the efficiency over all candidates of error structure. In order to appreciate the basic implications of each design criterion, these approaches are applied to two different experimental situations, D- and heteroscedastic G-optimalities.

• Communications for Statistical Applications and Methods
• /
• v.23 no.6
• /
• pp.575-585
• /
• 2016

Longitudinal studies repeatedly measure outcomes over time. Therefore, repeated measurements are serially correlated from same subject (within-subject variation) and there is also variation between subjects (between-subject variation). The serial correlation and the between-subject variation must be taken into account to make proper inference on covariate effects (Diggle et al., 2002). However, estimation of the covariance matrix is challenging because of many parameters and positive definiteness of the matrix. To overcome these limitations, we propose autoregressive moving average Cholesky decomposition (ARMACD) for the linear mixed models. The ARMACD allows a class of flexible, nonstationary, and heteroscedastic models that exploits the structure allowed by combining the AR and MA modeling of the random effects covariance matrix. We analyze a real dataset to illustrate our proposed methods.

• Journal of Korean Society of Transportation
• /
• v.21 no.4
• /
• pp.57-66
• /
• 2003

Because the Logit model easily calculates probabilities for choice alternatives and estimates parameters for explanatory variables, it is widely used as a traffic mode choice model. However, this model includes an assumption which is independently and identically distributed to the error component distribution of the mode choice utility function. This paper is a study on the estimation of the Heteroscedastic Logit Model. which mitigates this assumption. The purpose of this paper is to estimate a Logit model that more accurately reflects the mode choice behavior of passengers by resolving the homoscedasticity of the model choice utility error component. In order to do this, we introduced a scale factor that is directly related to the error component distribution of the model. This scale factor was defined so as to take into account the heteroscedasticity in the difference in travel time between using public transport and driving a car, and was used to estimate the travel time parameter. The results of the Logit Model estimation developed in this study show that Heteroscedastic Logit Models can realistically reflect the mode choice behavior of passengers, even if the difference in travel time between public and private transport remains the same as passenger travel time increases, by identifying the difference in mode choice probability of passengers for public transportation.

• The Korean Journal of Financial Management
• /
• v.23 no.2
• /
• pp.189-208
• /
• 2006

In this paper we are concerned with estimation of tail related risk measures for heteroscedastic financial time series and VaR limits that VaR tells us nothing about the potential size of the loss given. So we use GARCH-EVT model describing the tail of the conditional distribution for heteroscedastic financial series and adopt Expected Shortfall to overcome VaR limits. The main results can be summarized as follows. First, the distribution of stock return series is not normal but fat tail and heteroscedastic. When we calculate VaR under normal distribution we can ignore the heavy tails of the innovations or the stochastic nature of the volatility. Second, GARCH-EVT model is vindicated by the very satisfying overall performance in various backtesting experiments. Third, we founded the expected shortfall as an alternative risk measures.

• Communications for Statistical Applications and Methods
• /
• v.7 no.1
• /
• pp.87-96
• /
• 2000

Nonlinear time series prediction is derived and compared between statistic of modeling and neural network method. In particular mean squared errors of predication are obtained in generalized random coefficient model and generalized autoregressive conditional heteroscedastic model and compared with them by neural network forecasting.

• Journal of Korea Multimedia Society
• /
• v.17 no.12
• /
• pp.1446-1452
• /
• 2014

Existing LDA uses the transform matrix that maximizes distance between classes. So we have to convert from an image to one-dimensional vector as training vector. However, in 2D-LDA, we can directly use two-dimensional image itself as training matrix, so that the classification performance can be enhanced about 20% comparing LDA, since the training matrix preserves the spatial information of two-dimensional image. However 2D-LDA uses same calculation schema for transformation matrix and therefore both LDA and 2D-LDA has the heteroscedastic problem which means that the class classification cannot obtain beneficial information of spatial distances of class clusters since LDA uses only data correlation-based covariance matrix of the training data without any reference to distances between classes. In this paper, we propose a new method to apply training matrix of 2D-LDA by using WPS-LDA idea that calculates the reciprocal of distance between classes and apply this weight to between class scatter matrix. The experimental result shows that the discriminating power of proposed 2D-LDA with weighted between class scatter has been improved up to 2% than original 2D-LDA. This method has good performance, especially when the distance between two classes is very close and the dimension of projection axis is low.

• Smart Structures and Systems
• /
• v.32 no.4
• /
• pp.267-279
• /
• 2023

Data modelling and interpretation for structural health monitoring (SHM) field data are critical for evaluating structural performance and quantifying the vulnerability of infrastructure systems. In order to improve the data modelling accuracy, and extend the application range from data regression analysis to out-of-sample forecasting analysis, an improved most likely heteroscedastic Gaussian process (iMLHGP) methodology is proposed in this study by the incorporation of the outof-sample forecasting algorithm. The proposed iMLHGP method overcomes this limitation of constant variance of Gaussian process (GP), and can be used for estimating non-stationary typhoon-induced response statistics with high volatility. The first attempt at performing data regression and forecasting analysis on structural responses using the proposed iMLHGP method has been presented by applying it to real-world filed SHM data from an instrumented cable-stay bridge during typhoon events. Uncertainty quantification and correlation analysis were also carried out to investigate the influence of typhoons on bridge strain data. Results show that the iMLHGP method has high accuracy in both regression and out-of-sample forecasting. The iMLHGP framework takes both data heteroscedasticity and accurate analytical processing of noise variance (replace with a point estimation on the most likely value) into account to avoid the intensive computational effort. According to uncertainty quantification and correlation analysis results, the uncertainties of strain measurements are affected by both traffic and wind speed. The overall change of bridge strain is affected by temperature, and the local fluctuation is greatly affected by wind speed in typhoon conditions.

• Korean Management Science Review
• /
• v.11 no.2
• /
• pp.29-41
• /
• 1994

This paper analyses the stochastic nature of the Permanent Income Hypothesis (PIH) by specifying the variance-covariance structure of PIH based on Hall and Mishkin[3]. Maximum likelihood is employed to estimate the model by explicitely incorporating the heteroscedastic nature of the data into the likelihood. The data used are individual Korean household consumption and income data. The results indicate that the data are generally consistent with the Permanent Income Hypothesis, and about 11 percent of the total variation in consumption may be attributable to the excess sensitivity of consumption to income.