• Journal of Korean Society for Quality Management
• /
• v.23 no.4
• /
• pp.28-41
• /
• 1995

We consider the problem of identifying multiple outliers in linear model. The available regression diagnostic methods often do not succeed in detecting multiple outliers because of the masking and swamping effect. Recently, among the various robust estimator of reducing the effect of outliers, LMS(Least Meadian Square) estimator has been to be a suitable method proposed to expose outliers and leverage points. However, as you know it, the data analysis method with LMS estimator is to be taken the median of the squared residuals in the sample which is extracted the sample space. Then this model causes the trouble, for the number of the chosen sample is nCp, i.e. as the size of sample space n is increasing, the number is increasing fastly. And the covariance matrix may be the singular matrix, so that matrix is approching collinearity. Thus we propose a procedure ELMS for the resampling in LMS method and study the size of the effective elementary set in this algorithm.

• Journal of applied mathematics & informatics
• /
• v.28 no.3_4
• /
• pp.913-921
• /
• 2010

The derivative influence measure is adapted to the Cholesky decomposition of a covariance matrix. Formulas for the derivative influence of observations on the Cholesky root and the inverse Cholesky root of a sample covariance matrix are derived. It is easy to implement this influence diagnostic method for practical use. A numerical example is given for illustration.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.2
• /
• pp.78-88
• /
• 2009

In many face recognition problems, the number of available images is limited compared to the dimension of the input space which is usually equal to the number of pixels. This problem is called as the 'small sample size' problem and regularization methods are typically used to solve this problem in feature extraction methods such as LDA. By using regularization methods, the modified within class matrix becomes nonsingu1ar and LDA can be performed in its original form. However, in the process of adding a scaled version of the identity matrix to the original within scatter matrix, the scale factor should be set heuristically and the performance of the recognition system depends on highly the value of the scalar factor. By using the proposed resampling method, we can generate a set of images similar to but slightly different from the original image. With the increased number of images, the small sample size problem is alleviated and the classification performance increases. Unlike regularization method, the resampling method does not suffer from the heuristic setting of the parameter producing better performance.

• Journal of the Korean Statistical Society
• /
• v.22 no.2
• /
• pp.339-345
• /
• 1993

It is well known that distribution of functions of eigen values and vectors of a certain matrix plays an important role in multivariate analysis. This paper deals with the transformation of a correlation-type random matrix to its eigen values and vectors. Properties of the transformation are also considered. The results obtained are applied to express the joint distribution of eigen values and vectors of the correlation matrix when sample is taken from a m-variate spherical distribution.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.4
• /
• pp.1007-1012
• /
• 2003

A recursive procedure for finding the Cholesky root of the inverse of sample covariance matrix, leading to a direct solution for the inverse of a positive definite matrix, is developed using the likelihood equation for the maximum likelihood estimation of the Cholesky root under normality assumptions. An example of the Hilbert matrix is considered for an illustration of the procedure.

• Bulletin of the Korean Chemical Society
• /
• v.25 no.2
• /
• pp.233-236
• /
• 2004

Total impurity analysis of a primary standard solution is one of the essential procedures to determine an accurate concentration of the standard solution by the gravimetry. Bi impurity is determined in Pb standard solutions by inductively coupled plasma mass spectrometry (ICP-MS). The direct nebulization of the Pb standard solution produces a significant amount of the Pb matrix-induced molecular ions which give rise to a serious spectral interference to the Bi determination. In order to avoid the spectral interference from the interferent $^{208}PbH^+$, the hydride generation method is employed for the matrix separation. The Bi hydride vapor is generated by reaction of the sample solution with 1% sodium borohydride solution. The vapor is then directed by argon carrier gas into the ICP after separation from the mixture solution in a liquid-gas separator made of a polytetrafluoroethylene membrane tube. The presence of 1000 ${\mu}$g/mL Pb matrix caused reduction of the bismuthine generation efficiency by about 40%. The standard addition method is used to overcome the chemical interference from the Pb matrix. Optimum conditions are investigated for the hydride-generation ICPMS. The detection limit of this method is 0.5 pg/mL for the sample solutions containing 1000 ${\mu}$g/mL Pb matrix.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1993.06a
• /
• pp.206-209
• /
• 1993

The main purpose of this paper is to verify the effectiveness of a high resolution direction finding method, so called the‘averaged MUSIC’. This method uses a new sample array covariance matrix that consists of diagonal components obtained by taking averages of the diagonal component values of the sample covariance matrix for the MUSIC. The paper shows that the proposed method performs higher resolved direction-of-arrival estimation and better resolution probability than the MUSIC in such cases as low signal-to-noise ratio, when the number of sensors used is finite, based on the statistical analysis.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.22 no.2
• /
• pp.182-192
• /
• 1996

In this study, the relationship between wax matrix in lipstick and sweating was investigated by observing the change of size and shape of wax matrix with sweating by Scanning Electron Microscopy (SEM). For observation by SEM, a lipstick sample was frozen in liquid nitrogen, then the oil in the lipstick was extracted out in cold isopropanol($-70^{\circ}C$) for 1-3days. After isopropanol was evaporated, the sample was sputtered with gold, and examined by SEM. When examined the sweated sample by SEM, the change of wax matrix underneath the surface from fine, uniform structure to coarse, nonuniform structure was observed, which was resulted from the caking of surrounding wax matrix. That is, the oil underneath the surface was migrated to the surface of lipstick with sweating, consequently the wax matrix at that region was rearranged into the coarse matrix. In case of flamed lipstick, sweating was delayed and the wax matrix was much coarser than that of unflamed one. Its larger wax matrix at surface region was good for including oil. The effect of molding temperature on sweating was also studied. As the molding temperature was increased, sweating was greatly reduced and the size of wax matrix was increased. It was also found that sweating was influenced with the compatinility of wax and oil. A formula consisting of wax and oil which have good compatibility has a tendency of reduced sweating and increased size of wax matrix. When pigment was added to wax and oil. It was also found that sweating was influenced with the passage of time by observing a thick membrane of wax on surface of lipstick after a month from molding. In case of some lipsticks, the size of wax matrix was altered to bigger or smaller. In conclusion, the structure of wax matrix at the surface region of lipstick was changed with the process of foaming, molding temperature, compatibility of wax and oil, addition of pigment, and the passage of time. In most cases, as the size of wax matrix was increased, sweating was reduced and delayed.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2007.04a
• /
• pp.421-426
• /
• 2007

A parametric study to investigate the effects of composition variables on the glass transition and mechanical properties of CNT-based nanocomposites was performed using molecular dynamics simulations. In this study, matrix chain length and CNT length were chosen as the candidate characteristic parameters. In order to understand the effect of both parameters in detail, three sample sets having different chain lengths with the same CNT configuration and two sample sets having different CNT lengths with same chain length were prepared. Other parameters such as volume fraction and density were fixed to enable rigorous comparisons. Amorphous polyethylene is used as matrix polymer and (10,0) zigzag CNT is embedded into the matrix to reinforce polymer matrix. As a result, longer polymer chain length of matrix solely increased glass transition temperature but no reinforcing enhancement was observed. CNT length showed similar increase with little enhancement of elastic modulus. In addition to this, nanocomposites showed temperature-dependent elastic modulus jump passing thorough the glass transition region agrees well with experimental results.

• Analytical Science and Technology
• /
• v.20 no.1
• /
• pp.41-48
• /
• 2007

The determination method for the trace lead and cadmium in glass materials was developed by ICP-AES. The lead- and cadmium-DDTC chelates were formed by adding Na-DDTC (sodium diethyldithiocarbamate) as a chelating agent and extracted into MIBK (methyl-isobutyl ketone). Direct measuring of the trace lead and cadmium in glass was not possible because of the interferences of matrix elements as Na, Si, Mg and Ca of glasses unless solvent extraction of lead and cadmium was applied. This procedure was confirmed to be valid quantitative method of Pb and Cd analysis in glass did not influence of sample matrix, by applying it procedure to artifact sample and NIST SRM 1412.