• Geomechanics and Engineering
• /
• 제14권4호
• /
• pp.315-324
• /
• 2018

With rapid economic growth, numerous deep excavation projects for high-rise buildings and subway transportation networks have been constructed in the past two decades. Deep excavations particularly in thick deposits of soft clay may cause excessive ground movements and thus result in potential damage to adjacent buildings and supporting utilities. Extensive plane strain finite element analyses considering small strain effect have been carried out to examine the wall deflections for excavations in soft clay deposits supported by diaphragm walls and bracings. The excavation geometrical parameters, soil strength and stiffness properties, soil unit weight, the strut stiffness and wall stiffness were varied to study the wall deflection behaviour. Based on these results, a multivariate adaptive regression splines model was developed for estimating the maximum wall deflection. Parametric analyses were also performed to investigate the influence of the various design variables on wall deflections.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2001년도 춘계학술대회 논문집
• /
• pp.437-442
• /
• 2001

In Korea, collapse of railway slopes is frequently occurred due to tycoon or heavy rainfall doling rainy season and has been made personal and social damage greatly. In order to evaluate the stability of railway slopes under rainfall, explanatory variables and subordinate variables were selected for multivariate analysis. The sites which failure had occurred due to rainfall were investigated and critical rainfall was defined by the case that had high value of correlation factor after multivariate analyses for 121 cases had been executed. The maximum hourly rainfall during 24 hours before failure caused the collapse of railway embankment and the 0.3 square value of maximum hourly rainfall during 24 hours before failure caused the collapse of railway cut-slope, From the application to collapse examples, it is judged that critical rainfall curve will be used to estimate the stability of slopes.

• Computers and Concrete
• /
• 제16권4호
• /
• pp.569-585
• /
• 2015

Various computational tools are available for modeling highly nonlinear structural engineering problems that lack a precise analytical theory or understanding of the phenomena involved. This paper adopts a fairly simple nonparametric adaptive regression algorithm known as multivariate adaptive regression splines (MARS) to model the nonlinear interactions between variables. The MARS method makes no specific assumptions about the underlying functional relationship between the input variables and the response. Details of MARS methodology and its associated procedures are introduced first, followed by a number of examples including three practical structural engineering problems. These examples indicate that accuracy of the MARS prediction approach. Additionally, MARS is able to assess the relative importance of the designed variables. As MARS explicitly defines the intervals for the input variables, the model enables engineers to have an insight and understanding of where significant changes in the data may occur. An example is also presented to demonstrate how the MARS developed model can be used to carry out structural reliability analysis.

• Communications for Statistical Applications and Methods
• /
• 제25권6호
• /
• pp.605-618
• /
• 2018

Risk management has been a crucial part of the daily operations of the financial industry over the past two decades. Value at Risk (VaR), a quantitative measure introduced by JP Morgan in 1995, is the most popular and simplest quantitative measure of risk. VaR has been widely applied to the risk evaluation over all types of financial activities, including portfolio management and asset allocation. This paper uses the implementations of multivariate GARCH models and copula methods to illustrate the performance of a one-day-ahead VaR prediction modeling process for high-dimensional portfolios. Many factors, such as the interaction among included assets, are included in the modeling process. Additionally, empirical data analyses and backtesting results are demonstrated through a rolling analysis, which help capture the instability of parameter estimates. We find that our way of modeling is relatively robust and flexible.

• Asian Pacific Journal of Cancer Prevention
• /
• 제13권6호
• /
• pp.2511-2514
• /
• 2012

Objective: To evaluate the prognostic factors of gallbladder carcinoma. Methods: Presentation, operative data, complications, and survival outcome were examined for 132 gallbladder carcinoma patients who underwent gallbladder surgery in our unit during 2002-2007, and follow-up results were obtained from every patient for univariate and multivariate survival analysis. Results: The univariate analysis showed that gallbladder lesion history, tumor cell differentiation, Nevin staging, preoperative lymph node metastasis and the surgical approach significantly correlated with the prognosis of the patients (p<0.05). The results of the multivariate analysis (Cox regression) showed that gallbladder lesion history, Nevin staging and the surgical approach were independent predicators with relative risks of 6.9, 4.4, 2.8, respectively (p=0.002, 0.003, 0.008). Conclusion: Gallbladder lesion history, Nevin staging and the surgical approach are independent prognostic factors for gallbladder carcinoma, a rapidly fatal disease. Therefore, early diagnosis, anti-infective therapy and radical surgery are greatly needed to improve the prognosis of gallbladder carcinoma.

• International Journal of Contents
• /
• 제8권1호
• /
• pp.23-29
• /
• 2012

Multiple expression levels of genes obtained using time series microarray experiments have been exploited effectively to enhance understanding of a wide range of biological phenomena. However, the unique nature of microarray data is usually in the form of large matrices of expression genes with high dimensions. Among the huge number of genes presented in microarrays, only a small number of genes are expected to be effective for performing a certain task. Hence, discounting the majority of unaffected genes is the crucial goal of gene selection to improve accuracy for disease diagnosis. In this paper, a non-Gaussian weight matrix obtained from an incremental model is proposed to extract useful features of multivariate time series microarrays. The proposed method can automatically identify a small number of significant features via discovering hidden variables from a huge number of features. An unsupervised hierarchical clustering representative is then taken to evaluate the effectiveness of the proposed methodology. The proposed method achieves promising results based on predictive accuracy of clustering compared to existing methods of analysis. Furthermore, the proposed method offers a robust approach with low memory and computation costs.

• Bulletin of the Korean Chemical Society
• /
• 제31권4호
• /
• pp.825-828
• /
• 2010

To classify Glycyrrhiza species, samples of different species were analyzed by $^1H$ NMR-based metabolomics technique. Partial least squares discriminant analysis (PLS-DA) was used as the multivariate statistical analysis of the 1H NMR data sets. There was a clear separation between various Glycyrrhiza species in the PLS-DA derived score plots. The PLS-DA model was validated, and the key metabolites contributing to the separation in the score plots of various Glycyrrhiza species were lactic acid, alanine, arginine, proline, malic acid, asparagine, choline, glycine, glucose, sucrose, 4-hydroxy-phenylacetic acid, and formic acid. The compounds present at relatively high levels were glucose, and 4-hydroxyphenylacetic acid in G. glabra; lactic acid, alanine, and proline in G. inflata; and arginine, malic acid, and sucrose in G. uralensis. This is the first study to perform the global metabolomic profiling and differentiation of Glycyrrhiza species using $^1H$ NMR and multivariate statistical analysis.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.116.3-116
• /
• 2001

A diagnosis methodology for thickness quality in hot finishing mill is proposed based on multivariate statistical analysis. The thickness of hot strip is a key quality factor that is measured by x-ray thickness gauge. Currently, the thickness quality is guaranteed by upper and lower limit of thickness deviation from target thickness. But if any over-limit is occurred, there is no in-line method to identify the causes. In this paper, many parameters are extracted from the thickness deviation signal such as mean deviation(top, middle, tail), rms deviation(top, middle, tail) and peak deviation(top, middle, tail) as time domain parameters ...

• 한국관개배수논문집
• /
• 제5권1호
• /
• pp.75-82
• /
• 1998

A multivariate regression analysis for compression index and com- pressibility ratio of clayey soils in regard to some soil indices, i.e natural water content, Atterberg limits, and in-situ void ratio, was presented to estimate the primary consolidation s

• 한국컴퓨터정보학회논문지
• /
• 제5권3호
• /
• pp.142-150
• /
• 2000

범주형 다변량 데이터의 상관관계분석을 위하여 개발한 수량화이론 III류나 대응분석 등의 기법은 다차원 공간상에서 점간의 거리로써 두 요소집합간의 관련성을 설명하는 데 있어서 매우 유용하다. 본 연구에서는 상관관계분석을 위한 대응분석의 특성을 수량화이론 III류와 비교하여 설명하고 그 유용성을 논하기로 한다. 이 기법은 사회과학 분야의 상관관계분석에 널리 활용될 것으로 기대된다.