• The Korean Journal of Applied Statistics
• /
• v.29 no.3
• /
• pp.513-523
• /
• 2016

Fitting a mixture of multivariate skew normal distribution (MSNMix) with multiple skewness parameter vectors via EM algorithm often requires a highly expensive computational cost to calculate the moments and probabilities of multivariate truncated normal distribution in E-step. Subsequently, it is common to fit an asymmetric data set with MSNMix with a simple skewness parameter vector since it allows us to compute them in E-step in an univariate manner that guarantees a cheap computational cost. However, the adaptation of a simple skewness parameter is unrealistic in many situations. This paper proposes an approximate estimation for the MSNMix with multiple skewness parameter vectors that also allows us to treat them in an univariate manner. We additionally provide some experiments to show its effectiveness.

• Journal of Sensor Science and Technology
• /
• v.26 no.6
• /
• pp.371-378
• /
• 2017

An accurate and robust estimation of center of rotation (CoR) using optical marker trajectory is crucial in human biomechanics. In this regard, the performances of the two prevailing least-squares methods, the Gamage and Lasenby (GL) method, and the Chang and Pollard (CP) method, are verified in this paper. While both methods are sphere-fitting approaches in closed form and require no tuning parameters, they have not been thoroughly verified by comparison of their estimation accuracies. Furthermore, while for both methods, results for stationary CoR locations are presented, cases for perturbed CoR locations have not been investigated for any of them. In this paper, the estimation performances of the GL method and CP method are investigated by varying the range of motion (RoM) and noise amount, for both stationary and perturbed CoR locations. The difference in the estimation performance according to the variation in the amount of noise and RoM was clearly shown for both methods. However, the CP method outperformed the GL method, as seen in results from both the simulated and the experimental data. Particularly, when the RoM is small, the GL method failed to estimate the appropriate CoR while the CP method reasonably maintained the accuracy. In addition, the CP method showed a considerably better predictability in CoR estimation for the perturbed CoR location data than the GL method. Accordingly, it may be concluded that the CP method is more suitable than the GL method for CoR estimation when RoM is limited and CoR location is perturbed.

• Communications for Statistical Applications and Methods
• /
• v.24 no.4
• /
• pp.383-396
• /
• 2017

The model in our approach assumes that computer responses are a realization of a Gaussian processes superimposed on a regression model called a Gaussian process regression model (GPRM). Selecting a subset of variables or building a good reduced model in classical regression is an important process to identify variables influential to responses and for further analysis such as prediction or classification. One reason to select some variables in the prediction aspect is to prevent the over-fitting or under-fitting to data. The same reasoning and approach can be applicable to GPRM. However, only a few works on the variable selection in GPRM were done. In this paper, we propose a new algorithm to build a good prediction model among some GPRMs. It is a post-work of the algorithm that includes the Welch method suggested by previous researchers. The proposed algorithms select some non-zero regression coefficients (${\beta}^{\prime}s$) using forward and backward methods along with the Lasso guided approach. During this process, the fixed were covariance parameters (${\theta}^{\prime}s$) that were pre-selected by the Welch algorithm. We illustrated the superiority of our proposed models over the Welch method and non-selection models using four test functions and one real data example. Future extensions are also discussed.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.42 no.5
• /
• pp.133-143
• /
• 2000

A large scale field test of prefabricated vertical drains was performed to anayze the effect of parameters of the very soft clay at a test site. compression index and the coefficient of horizontal consolidation obtained by back-analysis of settlement data were compared with those obtained by means of laboratory tests. Hyperbolic method, Asaoka meoth and curve fitting method were used to compute final settlement of coefficient of consolidation. The relationships of settlement measurement(Sm) versus design settlement(St) and the measurement consolidation ratio(Um) versus design consolidation (Ut) were shown as Sm=(1.0~1.1) St , Um=(1.13~1.17) Ut at 1.0m spacing of drain and Sm=(0.7~0.8)St, Um= (0.92~0.99) Ut at 1.5 m spacing of drain, respectively . The relationships of the field compression index(CcField) and virgin compression index(vcc lab) were shown as Ccfield =(1.0~1.2)vcc lab . But it was nearly within the same range when considering the error factor with the determination method of virgin compression index and the prediction back-analysis of the settlement data was larger than the coefficient of vertical consolidation, and the ratio of consolidation coefficient (Ch/Cv) was Ch =(2.4~2.9) Cv , Ch=(3.4~4.2) Cv at 1.0m and 1.5m spacing of drain, respectively.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.28 no.1
• /
• pp.143-153
• /
• 2004

The Purpose of this study is provide fundamental data on a suitable sizes of ready-to-wear for women. The subjects of the survey were 300 female of 18 to 24 years old. They were measured direct anthropometry. Height, bust girth, waist girth and hip girth were grouped by KS interval. Data were analyzed by use of Cross Tab method. The results of this study were as follows. Body group of the anthropometric measurements were summed up as follows. Height was divided into 4 section(155-170), bust girth was divided into 7 section(76-94), waist girth was divided into 6 section(61-76) and hip girth was divided into 9 section(84-100). The characteristics of cross tab can be summarized as follows. Section with application ratio are 155-160cm of height, 82-85cm for bust girth, 64-70cm for waist girth and 86-94cm for hip girth. The sizes of high cover ratio of well-fitting jacket, dress, coat and one-piece was 17 sizes, well-fitting an upper garment and the whole body garment item except that was 16 sizes, and a lower garment was 21 sizes.

• Bulletin of the Korean Chemical Society
• /
• v.32 no.3
• /
• pp.863-866
• /
• 2011

The electrode reaction of $Eu^{3+}$ in a LiCl-KCl eutectic melt has been re-examined using cyclic voltammetry (CV). In this work, for the first time, the kinetic details of a $Eu^{3+}/Eu^{2+}$ redox system have been completely elucidated, along with the thermodynamic property, through a curve fitting applied to experimental CV data, which were obtained in a wide scan rate range of 0.5 to 10 V/s. The simulated results showed an excellent fit to all experimental CV data simultaneously, even though the curve fittings were performed within a large dynamic range of initial transfer coefficient values, formal potentials, and standard rate constants. As a result, a proper formal potential, transfer coefficient, and standard rate constant for the $Eu^{3+}/Eu^{2+}$ redox system were successfully extracted using the CV curve fitting.

• Journal of the Korean Home Economics Association
• /
• v.40 no.12
• /
• pp.119-129
• /
• 2002

This research aims at finding out situation on purchasing clothes for kids, size recognition and fitting to provide basic data in order to establish standard sizes for kids' clothes. A survey was conducted to 360 mothers who have children ranging from first to third grade in elementary school on clothes purchasing method, wearing situation, criteria in deciding size and fitting preference. In terms of data analysis, frequency, percentage, mean, $X^2$-test, ANOVA and Duncan-test were used via SPSS PC program. The result is as follows; 1. While the recognition on brand was high, the recognition on body size was low: as a result, it is believed that it would be effective to mark the weight along with the brand. 2. Many times, they would select one or two size bigger on all items and they would go for a regular or one that fits as far as filling is concerned showing that it is different from adult clothes. 3. As far as obese kids, they would choose based on the diameter, so the diameter item is right but they are too long in general. The length question should be given special attention when manufacturing clothes for obese kids.

• 한국지구물리탐사학회:학술대회논문집
• /
• 2003.11a
• /
• pp.390-397
• /
• 2003

A three-dimensional (3D) inversion technique has been developed for interpretation of magnetotelluric (MT) data. The inversion method is based on the linearized least-squares (Gauss-Newton) method with smoothness regularization. In addition to the underground 3D resistivity distribution, static shifts are also treated as unknown parameters in the inversion. The forward modeling is by the staggered-grid finite difference method. A Bayesian criterion ABle is applied to search the optimum trade-off among the minimization of the data misfit, model roughness and static shifts. The method has been applied to several MT datasets obtained at geothermal fields in Japan and other Asian countries. In this paper, two examples will be discussed: one is the data at the Ogiri geothermal area, southwestern Japan, and the other is at the Pohang low-enthalpy geothermal field, southeastern Korea. The inversion of the Ogiri data has been performed stably, resulting in a good fitting between the observed and computed apparent resistivities and phases. The recovered 3D resistivity structure is generally similar to the two-dimensional (2D) inversion models, although the deeper portion of the 3D model seems to be more realistic than that of the 2D model. The 3D model is also in a good agreement with the geological model of the geothermal reservoirs. 3D interpretation of the Pohang MT data is still preliminary. Although the fitting to the observed data is very good, the preliminary 3D model is not reliable enough because the station coverage is not sufficient for a 3D inversion.

• Journal of Electrical Engineering and Technology
• /
• v.8 no.4
• /
• pp.840-849
• /
• 2013

This paper presents a statistical approach to analyze electrical tree inception voltage, electrical tree breakdown voltage and tree breakdown time of unsaturated polyester resin subjected to AC voltage. The aim of this work was to show that Weibull and lognormal distribution may not be the most suitable distributions for analysis of electrical treeing data. In this paper, an investigation of statistical distributions of electrical tree inception voltage, electrical tree breakdown voltage and breakdown time data was performed on 108 leaf-like specimen samples. Revelations from the test results showed that Johnson SB distribution is the best fit for electrical tree inception voltage and tree breakdown time data while electrical tree breakdown voltage data is best suited with Wakeby distribution. The fitting step was performed by means of Anderson-Darling (AD) Goodness-of-fit test (GOF). Based on the fitting results of tree inception voltage, tree breakdown time and tree breakdown voltage data, Johnson SB and Wakeby exhibit the lowest error value respectively compared to Weibull and lognormal.

• Journal of Astronomy and Space Sciences
• /
• v.30 no.3
• /
• pp.145-152
• /
• 2013

Rotation-powered pulsars are excellent laboratories for studying particle acceleration as well as fundamental physics of strong gravity, strong magnetic fields and relativity. Particle acceleration and high-energy emission from the polar caps is expected to occur in connection with electron-positron pair cascades. I will review acceleration and gamma-ray emission from the pulsar polar cap and associated slot gap. Predictions of these models can be tested with the data set on pulsars collected by the Large Area Telescope on the Fermi Gamma-Ray Telescope over the last four years, using both detailed light curve fitting, population synthesis and phase-resolved spectroscopy.