• The Korean Journal of Applied Statistics
• /
• v.30 no.2
• /
• pp.259-269
• /
• 2017

Due to the nonnegativity of variance, most of nonparametric estimations of discontinuous variance function have used the Nadaraya-Watson estimation with residuals. By the modification of Chen et al. (2009) and Yu and Jones (2004), Huh (2014, 2016a) proposed the estimators of the log-variance function instead of the variance function using the local linear estimator which has no boundary effect. Huh (2016b) estimated the variance function using the adjusted squared residuals by the estimated jump size in the discontinuous variance function. In this paper, we propose an estimator of the discontinuous log-variance function using the local linear estimator with the adjusted log-squared residuals by the estimated jump size of log-variance function like Huh (2016b). The numerical work demonstrates the performance of the proposed method with simulated and real examples.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.9 no.4
• /
• pp.351-358
• /
• 2016

Science and technology is developing rapidly as more powerful software with the rapid development of software testing and reliability assessment by the difficulty increases with the complexity of the software features of the larger increases NHPP software reliability models for failure analysis can have, in the literature, exhibit either constant, monotonic increasing or monotonic decreasing failure occurrence rates per fault. In this paper, finite failure NHPP models that assuming the expected value of the defect and infinite failures NHPP models that repairing software failure point in time reflects the situation, were presented for comparing property. Commonly used in the field of software reliability based on Erlang distribution software reliability model finite failures and infinite failures were presented for performance comparative evaluation problem. As a result, finite failure model is better than infinite failure model effectively. The parameters estimation using maximum likelihood estimation in the course of this study was conducted. As the results of this research, software developers to identify software failure property be able to help is concluded.

• Journal of Digital Convergence
• /
• v.13 no.12
• /
• pp.143-149
• /
• 2015

This study aims to analyze the predict capability of some of the popular software NHPP reliability models(Goel-Okumo model, delayed S-shaped reliability model and Rayleigh distribution model). The predict capability analysis will be on two key factors, one pertaining to the degree of fitment on available failure data and the other for its prediction capability. Estimation of parameters for each model was used maximum likelihood estimation using first 80% of the failure data. Comparison of predict capability of models selected by validating against the last 20% of the available failure data. Through this study, findings can be used as priori information for the administrator to analyze the failure of software.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.25 no.6
• /
• pp.422-432
• /
• 2013

A System Identification method to develop parametric models linking free surface elevation and wave pressure is presented and two models are built allowing for either wave pressure or free surface elevation simulation. Linear, time invariant model structures with static nonlinearities are assumed and solutions are sought in a form of autoregressive model with extra input (ARX). An arbitrary chosen free-surface elevation and wave pressure dataset is used for estimation of the models, which are subsequently verified against datasets with similar pressure gauge depth but different free-surface elevation spectra due to different meteorological conditions. It is shown that free-surface simulation using System Identification methods can perform better than traditional linear transfer function derived from linear wave theory (LTF), while wave pressure simulation quality using presented methods is generally similar to that obtained with corrected LTF.

• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.733-745
• /
• 2017

Local composite quantile regression is a useful non-parametric regression method widely used for its high efficiency. Data smoothing methods using kernel are typically used in the estimation process with performances that rely largely on the smoothing parameter rather than the kernel. However, $L_2$-norm is generally used as criterion to estimate the performance of the regression function. In addition, many studies have been conducted on the selection of smoothing parameters that minimize mean square error (MSE) or mean integrated square error (MISE). In this paper, we explored the optimality of selecting smoothing parameters that determine the performance of non-parametric regression models using local linear composite quantile regression. As evaluation criteria for the choice of smoothing parameter, we used mean absolute error (MAE) and mean integrated absolute error (MIAE), which have not been researched extensively due to mathematical difficulties. We proved the uniqueness of the optimal smoothing parameter based on MAE and MIAE. Furthermore, we compared the optimal smoothing parameter based on the proposed criteria (MAE and MIAE) with existing criteria (MSE and MISE). In this process, the properties of the proposed method were investigated through simulation studies in various situations.

• Journal of Korean Society of Forest Science
• /
• v.110 no.3
• /
• pp.322-340
• /
• 2021

Due to climate change and its consequential spring temperature rise, flowering time of Robinia pseudoacacia has advanced and a simultaneous blooming phenomenon occurred in different regions in South Korea. These changes in flowering time became a major crisis in the domestic beekeeping industry and the demand for accurate prediction of flowering time for R. pseudoacacia is increasing. In this study, we developed and compared performance of four different models predicting flowering time of R. pseudoacacia for the entire country: a Single Model for the country (SM), Modified Single Model (MSM) using correction factors derived from SM, Group Model (GM) estimating parameters for each region, and Local Model (LM) estimating parameters for each site. To achieve this goal, the bloom date data observed at 26 points across the country for the past 12 years (2006-2017) and daily temperature data were used. As a result, bloom dates for the north central region, where spring temperature increase was more than two-fold higher than southern regions, have advanced and the differences compared with the southwest region decreased by 0.7098 days per year (p-value=0.0417). Model comparisons showed MSM and LM performed better than the other models, as shown by 24% and 15% lower RMSE than SM, respectively. Furthermore, validation with 16 additional sites for 4 years revealed co-krigging of LM showed better performance than expansion of MSM for the entire nation (RMSE: p-value=0.0118, Bias: p-value=0.0471). This study improved predictions of bloom dates for R. pseudoacacia and proposed methods for reliable expansion to the entire nation.

• Journal of Animal Science and Technology
• /
• v.48 no.3
• /
• pp.337-344
• /
• 2006

The objective of this study was to compare of genetic parameter estimates on carcass traits of Hanwoo(Korean Cattle) according to modeling with Gibbs sampler and MTDFREML. The data set consisted of 1,941 cattle records with 23,058 animals in pedigree files at Hanwoo Improvement Center. The variance and covariance among carcass traits were estimated via Gibbs sampler and MTDFREML algorithms. The carcass traits considered in this study were longissimus dorsi area, backfat thickness, and marbling score. Genetic parameter estimates using Gibbs sampler and MTDFREML from single-trait analysis were similar with those from multiple-trait analysis. The estimated heritabilities using Gibbs sampler were .52～.54, .54 ～.59, and .42～.44 for carcass traits. The estimated heritabilities using MTDFREML were .41, .52～.53, and .31～.32 for carcass traits. The estimated genetic correlation using Gibbs sampler and MTDFREML of LDA between BF and MS were negatively correlated as .34～.36, .23～.37. Otherwise, genetic correlation between BF and MS was positive genetic correlation as .36～.44. The correlations of breeding value for marbling score between via MTDFREML and via Gibbs sampler were 0.989, 0.996 and 0.985 for LDA, BF and MS respectively.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.8 no.5
• /
• pp.345-353
• /
• 2015

There are many software reliability models that are based on the times of occurrences of errors in the debugging of software. It is shown that it is possible to do parameter inference for software reliability models based on finite failure model and non-homogeneous Poisson Processes (NHPP). For someone making a decision to market software, the conditional failure rate is an important variables. In this case, finite failure model are used in a wide variety of practical situations. Their use in characterization problems, detection of outlier, linear estimation, study of system reliability, life-testing, survival analysis, data compression and many other fields can be seen from the many study. Statistical process control (SPC) can monitor the forecasting of software failure and thereby contribute significantly to the improvement of software reliability. Control charts are widely used for software process control in the software industry. In this paper, proposed a control mechanism based on NHPP using mean value function of polynomial hazard function.

• KDI Journal of Economic Policy
• /
• v.32 no.3
• /
• pp.71-99
• /
• 2010

The concept of CoVaR introduced by Adrian and Brunnermeier (2009) is a useful tool to measure the risk spillover effect. It can capture the risk contribution of each institution to overall systemic risk. While Adrian and Brunnermeier rely on the quantile regression method in the estimation of CoVaR, we propose a new estimation method using parametric distribution functions such as bivariate normal and $S_U$-normal distribution functions. Based on our estimates of CoVaR for Korean banking industry, we investigate the practical usefulness of CoVaR for a systemic risk measure, and compare the estimation performance of each model. Empirical results show that bank makes a positive contribution to system risk. We also find that quantile regression and normal distribution models tend to considerably underestimate the CoVaR (in absolute value) compared to $S_U$-normal distribution model, and this underestimation becomes serious when the crisis in a financial system is assumed.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.4
• /
• pp.641-648
• /
• 2022

In this study, the performance of the NHPP software reliability model with exponential distribution (Exponential Basic, Inverse Exponential, Lindley, Rayleigh) characteristics was comparatively analyzed, and based on this, the optimal reliability model was also presented. To analyze the software failure phenomenon, the failure time data collected during system operation was used, and the parameter estimation was solved by applying the maximum likelihood estimation method (MLE). Through various comparative analysis (mean square error analysis, true value predictive power analysis of average value function, strength function evaluation, and reliability evaluation applied with mission time), it was found that the Lindley model was an efficient model with the best performance. Through this study, the reliability performance of the distribution with the characteristic of the exponential form, which has no existing research case, was newly identified, and through this, basic design data that software developers could use in the initial stage can be presented.