• Journal of the Korean Data and Information Science Society
• /
• v.17 no.2
• /
• pp.467-474
• /
• 2006

In this paper we present a weighted support vector machine(SVM) and a weighted least squares support vector machine(LS-SVM) for the prediction in the heteroscedastic regression model. By adding weights to standard SVM and LS-SVM the better fitting ability can be achieved when errors are heteroscedastic. In the numerical studies, we illustrate the prediction performance of the proposed procedure by comparing with the procedure which combines standard SVM and LS-SVM and wild bootstrap for the prediction.

• Communications for Statistical Applications and Methods
• /
• v.6 no.2
• /
• pp.585-594
• /
• 1999

Let {$X_{nj}$, 1$\leq$j$\leq$n,j$\geq$1} be a triangular array of random variables which are neither independent nor identically distributed. The almost sure convergences of randomly weighted partial sums of the form $$\sum_n^{j=1}$$ $W_{nj}$$X_{nj} are studied where {Wnj 1$\leq$j$\leq$n, j$\geq$1} is a triangular array of random weights. Application regarding the Efron bootstrap is also introduced.

• Journal of the Korean Data and Information Science Society
• /
• v.15 no.1
• /
• pp.227-237
• /
• 2004

In this paper we presented a Bayesian inference approach for estimating the location and scale parameters of the lognormal distribution using iterative Gibbs sampling algorithm. We also presented estimation of location parameter by two non iterative methods, importance sampling and weighted bootstrap assuming scale parameter as known. The estimates by non iterative techniques do not depend on the specification of hyper parameters which is optimal from the Bayesian point of view. The estimates obtained by more sophisticated Gibbs sampler vary slightly with the choices of hyper parameters. The objective of this paper is to illustrate these tools in a simpler setup which may be essential in more complicated situations.

• The Korean Journal of Applied Statistics
• /
• v.17 no.3
• /
• pp.489-498
• /
• 2004

In this paper, a new form of linear models referred to as generalized weighted linear models is proposed. The proposed models assume that the relationship between the response variable and explanatory variables can be modelled by a distribution function of the response mean and a weighted linear combination of distribution functions of covariates. This form addresses a structural problem of the link function in the generalized linear models in which the parameter space may not be consistent with the space derived from linear predictors. The maximum likelihood estimation with Lagrange's undetermined multipliers is used to estimate the parameters and resampling method is applied to compute confidence intervals and to test hypotheses.

• Communications for Statistical Applications and Methods
• /
• v.26 no.3
• /
• pp.261-272
• /
• 2019

In this paper, we propose a procedure to build a prediction interval of the sum of dependent binary random variables over a graph to account for the dependence among binary variables. Our main interest is to find a prediction interval of the weighted sum of dependent binary random variables indexed by a graph. This problem is motivated by the prediction problem of various elections including Korean National Assembly and US presidential election. Traditional and popular approaches to construct the prediction interval of the seats won by major parties are normal approximation by the CLT and Monte Carlo method by generating many independent Bernoulli random variables assuming that those binary random variables are independent and the success probabilities are known constants. However, in practice, the survey results (also the exit polls) on the election are random and hardly independent to each other. They are more often spatially correlated random variables. To take this into account, we suggest a spatial auto-regressive (AR) model for the surveyed success probabilities, and propose a residual based bootstrap procedure to construct the prediction interval of the sum of the binary outcomes. Finally, we apply the procedure to building the prediction intervals of the number of legislative seats won by each party from the exit poll data in the $19^{th}$ and $20^{th}$ Korea National Assembly elections.

• Proceedings of the Korean Society for Quality Management Conference
• /
• 2006.11a
• /
• pp.199-204
• /
• 2006

ANOVA is widely used for measurement system analysis. It assumes that the measurement error is normally distributed, which may not be seen in some industrial cases. In this study, the estimates of the measurement system variability and PTR (precision-to-tolerance ratio) are obtained by using weighted standard deviation for the case where the measurement error is non-normally distributed. The Standard Bootstrap method is used for estimating confidence intervals of measurement system variability and PTR. The point and confidence interval estimates for the cases with normally distributed measurement error are compared to those with non-normally distributed measurement errors through computer simulation.

• Journal of Korean Society for Quality Management
• /
• v.35 no.1
• /
• pp.10-19
• /
• 2007

ANOVA is widely, used for measurement system analysis. It assumes that the measurement error is normally distributed, which nay not be seen in some industrial cases. In this study the estimates of the measurement system variability and PTR (precision-to-tolerance ratio) are obtained by using weighted standard deviation for the case where the measurement error is non-normally distributed. The Standard Bootstrap method is used foy estimating confidence intervals of measurement system variability and PTR. The point and confidence interval estimates for the cases with normally distributed measurement error are compared to those with non-normally distributed measurement errors through computer simulation.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.30 no.4
• /
• pp.302-310
• /
• 2010

In this paper, we studied Bagging neural network for predicting defect size of steam generator(SG) tube in nuclear power plant. Bagging is a method for creating an ensemble of estimator based on bootstrap sampling. For predicting defect size of SG tube, we first generated eddy current testing signals for 4 defect patterns of SG tube with various widths and depths. Then, we constructed single neural network(SNN) and Bagging neural network(BNN) to estimate width and depth of each defect. The estimation performance of SNN and BNN were measured by means of peak error. According to our experiment result, average peak error of SNN and BNN for estimating defect depth were 0.117 and 0.089mm, respectively. Also, in the case of estimating defect width, average peak error of SNN and BNN were 0.494 and 0.306mm, respectively. This shows that the estimation performance of BNN is superior to that of SNN.

• Journal of Korea Port Economic Association
• /
• v.36 no.2
• /
• pp.109-124
• /
• 2020

Using panel data on freight rates and ship prices in the dry freighter market from January 2015 to December 2019, this study investigates the characteristics of shipping industry fluctuations. The analysis aims at two aspects of academic contribution. First, this study analyzes the relationship between shipping indicators and ship price based on separate dry-bulk ships, while the previous research considered the overall shipping index and weighted average ship prices. Second, the VAR model for the causality test is extended to a heterogeneous mixed panel model capable of limiting coefficients. There is a peak estimated by removing the cross-correlation problem, which is mainly raised in panel data analysis, using bootstrap estimation and solving the problem of information loss due to differences in non-stationary data. An empirical investigation of the causal relationship between economic fluctuations and ship price shows that the effect on the ship price from the freight is significant at the 1% level. This implies that there is a one-way relationship with demand in the shipping industry rather than a bilateral relationship.

• Journal of Korea Water Resources Association
• /
• v.43 no.2
• /
• pp.187-199
• /
• 2010

This study evaluated applicability and confidence of probability rainfalls estimated by the non-stationary rainfall frequency analysis which was recently developed. Using rainfall data at 4 sites which have an obvious increasing trend in observations, we estimated 3 type probability rainfalls; probability rainfalls from stationary rainfall frequency analysis using data from 1973-1997, probability rainfalls from stationary rainfall frequency analysis using data from 1973-2006, probability rainfalls from non-stationary rainfall frequency analysis assuming that the current year is 1997 and the target year is 2006. Based on the comparison of residuals from 3 probability rainfalls, the non-stationary rainfall frequency analysis provided more effective and well-directed estimates of probability rainfalls in the target year. Using Bootstrap resampling, this study also evaluated the parameter estimation methods for the non-stationary rainfall frequency analysis based on confidence intervals. The confidence interval length estimated by the maximum likelihood estimation (MLE) is narrower than the probability weighted moments (PWM). The results indicated that MLE provides more proper confidence than PWM for non-stationary probability rainfalls.