• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2004.10a
• /
• pp.578-581
• /
• 2004

본 연구에서는 수요가 정규분포의 형태를 따르는 N차 시리얼(serial) 시스템을 대상으로 한다. 최상위 노드는 상위의 공급자로부터 받고자 하는 물량을 제한 없이 받을 수 있으며 하위 노드로 이러한 물량을 공급하게 된다. 최하위 노드에서는 고객의 직접적인 수요가 발생하고 만족시키지 못한 수요는 다음 기간으로 이월된다. 이러한 환경 하에서 전체시스템에서 발생하는 재고유지 비용(holding cost)과 재고이월 비용(backorder cost)의 합을 최소화하는 각 노드별 최적의 기초 재고 수준(base stock level)을 결정하는 문제를 다룬다. 본 논문에서는 모의실험과 계층 재고(echelon stock)의 개념을 통해 수요 분포 내의 적절한 분위수(quantile)를 결정하는 접근방법으로 각 노드의 기초 재고 수준을 구하는 방안을 제시하고자 한다.

• Korea Real Estate Review
• /
• v.28 no.1
• /
• pp.79-90
• /
• 2018

The purpose of this study was to estimate the pricing factors of residential lands in new cities by estimating the pricing model of residential lands. For this purpose, hedonic equations for each quantile of the conditional distribution of land prices were estimated using quantile regression methods and the sale price date of Jangyu New Town in Gimhae. In this study, a quantile regression method that models the relation between a set of explanatory variables and each quantile of land price was adopted. As a result, the differences in the effects of the characteristics by price quantile were confirmed. The number of years that elapsed after the completion of land construction is the quadratic effect in the model because its impact may give rise to a non-linear price pattern. Age appears to decrease the price until certain years after the construction, and increases the price afterward. In the estimation of the quantile regression, land age appears to have a statistically significant impact on land price at the traditional level, and the turning point appears to be shorter for the low quantiles than for the higher quantiles. The positive effects of the use of land for commercial and residential purposes were found to be the biggest. Land demand is preferred if there are more than two roads on the ground. In this case, the amount of sunshine will improve. It appears that the shape of a square wave is preferred to a free-looking land. This is because the square land is favorable for development. The variables of the land used for commercial and residential purposes have a greater impact on low-priced residential lands. This is because such lands tend to be mostly used for rental housing and have different characteristics from residential houses. Residential land prices have different characteristics depending on the price level, and it is necessary to consider this in the evaluation of the collateral value and the drafting of real estate policy.

• Journal of the Korea Society for Simulation
• /
• v.15 no.3
• /
• pp.39-47
• /
• 2006

In most of supply chain planning practices, the estimated demands, which are forecasted for each individual period in a forecasting window, are regarded as deterministic. But, in reality, the forecasted demands for the periods of a given horizon are stochastically distributed. Instead of using a safety stock, this study considers a direct control of service level by choosing the demand used in planning from the distributed forecasted demand values for the corresponding period. Using the demand quantile and echelon stock concept, we propose a simple but efficient heuristic algorithm for multi-echelon serial systems under service level constraints. Through a comprehensive simulation study, the proposed algorithm was shown to be very accurate compared with the optimal solutions.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.342-344
• /
• 2018

This paper shows the possibility to substitute statistical methods such as mean imputation, correlation coefficient analysis, graph correlation analysis for the proposed algorithm, and replace statistician for processing various abnormal data measured in the water treatment process with it. In addition, this study aims to model a data-filtering system based on a recent fractile pattern and a deep learning-based LSTM algorithm in order to improve the reliability and validation of the algorithm, using the open-sourced libraries such as KERAS, THEANO, TENSORFLOW, etc.

• The Journal of Engineering Geology
• /
• v.27 no.4
• /
• pp.367-376
• /
• 2017

Recently, tunneling under the seabed is becoming increasingly common in many countries. In Korea, there are proposals to tunnel from the mainland to Jeju Island. Safe construction requires geologic structures such as faults to be characterized during the design and construction phase; however, unlike on land, such structures are difficult to survey seabed. This study aims to develop an algorithm that uses geostatistics to automatically derive large-scale geological structures on the seabed. The most important considerations in this method are the optimal size of the moving window, the optimal type of spatial statistics, and determination of the optimal percentile standard. Finally, the optimal analysis algorithm was developed using the R program, which comprehensibly presents variations in spatial statistics. The program allows the type and percentile standard of spatial statistics to be specified by the user, thus enabling an analysis of the geological structure according to variations in spatial statistics. The geotechnical defense-training algorithm shows that a large, linear geological lineament is best visualized using a $3{\times}3$ moving window and a 10% upper standard based on the moving variance value and fractile. In particular, setting the fractile criterion to the upper 0.5% almost entirely eliminates the error values from the contour image.

• The Korean Journal of Applied Statistics
• /
• v.29 no.1
• /
• pp.75-83
• /
• 2016

We propose methodology to analyze the dynamic mechanisms of financial market contagion under market integration using a biological contagion analytical approach. We employ U-statistic to measure market integration, and a dynamic model based on an error correction mechanism (single equation error correction model) and latent factor model to examine market contagion. We also use quantile regression and Wald-Wolfowitz runs test to test market contagion. This methodology is designed to effectively handle heteroscedasticity and correlated errors. Our simulation results show that the single equation error correction model fits well with the linear regression model with a stationary predictor and correlated errors.

• Journal of Korea Technology Innovation Society
• /
• v.14 no.2
• /
• pp.373-404
• /
• 2011

The purpose of research is to reveal the effect of innovation to enterprises' economic performance. The kind of this study has begun since 1960s and lively progressed then. The fmal theoretical result of the effect of innovation to the performance came positive in compare to the mixed results came out in empirical analysis. There are several reason why empirical results are different to the theoretical results. However the major factor is that of using imperfect statistics and inappropriateness of analysis method. This study used a population (1990~2008) provided from Korean Intellectual Property Office, KIPO for patent and also used a population (1990~2008) provided from Korea Investors Service, KIS for research and development. The contribution of this study is enormous statistical analysis. This study used principal component analysis made innovativeness index for appropriate index sampling, and made effort to minimize the error by using appropriate quantile regression for both to panel analysis and rapidly developed company analysis. Dividing the final results into two parts, the growth and the profit, the effect of technological innovation to the firm's growth is not significant to the panel analysis but heavily significant to the upper 10% of high growth firm. By classifying large company and small and medium enterprise, it is significant to upper 10% of high growth firm for large company and generally significant to small and medium enterprise. But for both lower 10% of low growth firms and 25% of low ranking firms are negatively effected, and for high growth firms larger than the medians are positively effected. Especially for upper 10% of high growth firms are mostly effected. It is more effective to the profitability than the growth. The effect to the profit for every enterprises are not significant, but effected significant to the larger enterprises than 25% of low ranking enterprises especially most effective to the upper 10% of high-profit enterprises. The analysis for the large company, it was significant and positively effected to the upper 10% of high profit enterprises and 25% of low ranking enterprises, but the negatively effected for the low-profit enterprises. For the small and medium enterprises, it is negatively effected for both 10% of low ranking enterprises and 25% of low ranking enterprises. However it is positively effective and significant for the high ranking enterprises than median, especially for those high growth firms. It is meaningful to recognize significancy by quantile, but more implicative result is to finding more effectiveness to the small and medium enterprises than to the large company.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.621-636
• /
• 2018

Research on the application of informative sampling technique has been conducted in order to reduce the influence of non-response. Chung and Shin (Korean Journal of Applied Statistics, 30, 993-1004, 2017) showed that the estimation accuracy improved when using exponential response rate information for the parameter estimation if the distribution of errors included in the super population model follows normal distribution. However this method divides the stratum into equally spaced substrata to obtain the sample weight of the informative sampling technique and shows that the accuracy of the estimation improves as the number of substrata increases. In this study, with the given number of total sample size, the optimal substratum boundary points are calculated using equal space, quantile, and LH algorithm; consequently, the results using those methods are compared through simulation. We also studied the criteria to determine the number of substrata and substratum boundaries that can be used in practice with various types of auxiliary variable distributions.

• The Korean Journal of Applied Statistics
• /
• v.26 no.1
• /
• pp.37-47
• /
• 2013

A Q-Q plot is an effective and convenient graphical method to assess a distributional assumption of data. The primary step in the construction of a Q-Q plot is to obtain a closed-form expression to represent the relation between observed quantiles and theoretical quantiles to be plotted in order that the points fall near the line y = a + bx. In this paper, we introduce a Q-Q plot to assess goodness-of-fit for inverse Gaussian distribution. The procedure is based on the distributional result that a transformed random variable $Y={\mid}\sqrt{\lambda}(X-{\mu})/{\mu}\sqrt{X}{\mid}$ follows a half-normal distribution with mean 0 and variance 1 when a random variable X has an inverse Gaussian distribution with location parameter ${\mu}$ and scale parameter ${\lambda}$. Simulations are performed to provide a guideline to interpret the pattern of points on the proposed inverse Gaussian Q-Q plot. An illustrative example is provided to show the usefulness of the inverse Gaussian Q-Q plot.

• The Korean Journal of Applied Statistics
• /
• v.29 no.4
• /
• pp.689-697
• /
• 2016

The most useful method for financial market risk management may be Value at Risk (VaR) which estimates the maximum loss amount statistically. The VaR is used as a risk measure for one industry. Many real cases estimate VaRs for many industries or nationwide industries; consequently, it is necessary to estimate the VaR for multivariate distributions when a specific portfolio is established. In this paper, the multivariate quantile vector is proposed to estimate VaR for multivariate distribution, and the Vector at Risk for multivariate space is defined based on the quantile vector. When a weight vector for a specific portfolio is given, one point among Vector at Risk could be found as the best VaR which is called as an alternative VaR. The alternative VaR proposed in this work is compared with the VaR of Morgan with bivariate and trivariate examples; in addition, some properties of the alternative VaR are also explored.