• Communications for Statistical Applications and Methods
• /
• v.7 no.2
• /
• pp.574-574
• /
• 2000

Consider the problem of estimating regression function from a set of data which is contaminated by a long-tailed error distribution. The linear smoother is a kind of a local weighted average of response, so it is not robust against outliers. The kernel M-smoother and the lowess attain robustness against outliers by down-weighting outliers. However, the kernel M-smoother and the lowess requires the iteration for computing the robustness weights, and as Wang and Scott(1994) pointed out, the requirement of iteration is not a desirable property. In this article, we propose the robust nonparametic regression method which does not require the iteration. Robustness can be achieved not only by down-weighting outliers but also by transforming outliers. The rank transformation is a simple procedure where the data are replaced by their corresponding ranks. Iman and Conover(1979) showed the fact that the rank transformation is a robust and powerful procedure in the linear regression. In this paper, we show that we can also use the rank transformation to nonparametric regression to achieve the robustness.

• Communications for Statistical Applications and Methods
• /
• v.7 no.2
• /
• pp.575-583
• /
• 2000

Consider the problem of estimating regression function from a set of data which is contaminated by a long-tailed error distribution. The linear smoother is a kind of a local weighted average of response, so it is not robust against outliers. The kernel M-smoother and the lowess attain robustness against outliers by down-weighting outliers. However, the kernel M-smoother and the lowess requires the iteration for computing the robustness weights, and as Wang and Scott(1994) pointed out, the requirement of iteration is not a desirable property. In this article, we propose the robust nonparametic regression method which does not require the iteration. Robustness can be achieved not only by down-weighting outliers but also by transforming outliers. The rank transformation is a simple procedure where the data are replaced by their corresponding ranks. Iman and Conover(1979) showed the fact that the rank transformation is a robust and powerful procedure in the linear regression. In this paper, we show that we can also use the rank transformation to nonparametric regression to achieve the robustness.

• The Journal of Korea Robotics Society
• /
• v.8 no.4
• /
• pp.292-297
• /
• 2013

This paper presents a new method of kinematic modeling for autonomous bicycle by using the differential motion transformation. Kinematic model is indispensable to trajectory planning and control for an autonomous mobile robot. The conventional methods of kinematic modeling for an autonomous bicycle depend on intuition by geometry. On the contrary, the proposed method in this paper is based on the systematic differential motion transformation, thus applicable to various types of autonomous bicycles. The differential motion transformation gives Jacobian between two coordinate frames and the velocity kinematics as a result.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.38 no.5
• /
• pp.348-354
• /
• 1989

The metallurgical transformation ratios of metals have been evaluated by the eddy current method. Eddy current probes have been designed and applied to various standard samples. The equivalent impedance estimated from the eddy current measurement is closely related to the variation of transformation ratios in metals. It has been found that the variation of the impedance is relatively large in the samples where the transformation ratios of the samples are small. Both reactance and resistance increase when the transformation ratio is large. Sensitivities of eddy current test is increased with increasing frequencies and by inserting ferrite core the solenoid eddy probe. Our results have shown the implementation of eddy current measurement to the one-line nondestructive testing of metals.

• Asian Journal of Innovation and Policy
• /
• v.11 no.2
• /
• pp.222-237
• /
• 2022

As the COVID-19 pandemic crisis left developing countries with economic setbacks, it is high time to highlight that innovative technologies lead the digital economy. The big powers including the United States and China are already implementing industrial policies that involve large-scale fiscal expenditures to secure the lives and safety of their people. To prepare for the future up to 2025, this paper reflects opinions of industry-academia-research experts regarding changes in the external environment and industry trends. By reflecting results of focus group interviews and changes in the external environment and industry trends, a new high-level 5X strategy (Digital Transformation, Energy Transformation, Bio Health Transformation, Supply Chain Transformation, and Research Transformation) to solve national tasks required for the existing ten policy demand fields and ten agenda during lower-level policy implementation stages were derived.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.1E
• /
• pp.20-24
• /
• 2001

We propose the computation reduction method of real root method that is mainly used in the CELP(Code Excited Linear Prediction) vocoder. The real root method is that if polynomial equations have the real roots, we are able to find those and transform them into LSF[1]. However, this method takes much time to compute, because the root searching is processed sequentially in frequency region. But, the important characteristic of LSF is that most of coefficients are occurred in specific frequency region. So, the searching frequency region is ordered by each coefficient's distribution. And coefficients are searched in ordered frequency region. Transformation time can be reduced by this method than the sequential searching method in frequency region. When we compare this proposed method with the conventional real root method, the experimental result is that the searching time was reduced about 46% in average.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.1444-1447
• /
• 1997

In this paper, the problen of dseigning a H.inf. controller is considered, where the controller is realized through digital equipment. We show that the existing discrete-time controller design method can be improved by usign the inveres bilinear transformation. The usefulness of the given method is confirmed by simulation.

• Tunnel and Underground Space
• /
• v.3 no.2
• /
• pp.163-166
• /
• 1993

To improve the applicability of the Ho-Brown's empirical failure theory to field problems, transformation of failure coefficients was suggested in this study. This method was tested on the published data and good results were obtained. It seemed that this method be more reasonable than traditional method of modifying the coefficients by RMR or Q values.

• Structural Engineering and Mechanics
• /
• v.45 no.5
• /
• pp.631-641
• /
• 2013

In this paper, the problem of axisymmetric deformation of the circular membrane fixed at its perimeter under the action of uniformly-distributed loads was resolved by exactly using power series method, and the solution of the problem was presented. An investigation into the so-called Hencky transformation was carried out, based on the solution presented here. The results obtained here indicate that the well-known Hencky solution is, without doubt, correct, but in the published papers the statement about its derivation is incorrect, and the so-called Hencky transformation is invalid and hence may not be extended to use as a general mathematical method.

• Economic and Environmental Geology
• /
• v.18 no.3
• /
• pp.217-224
• /
• 1985

A Householder's transformation is applied to the resistivity inversion problem. The conventional resistivity inversion method is sometimes numerically unstable in interpreting a resistivity sounding data because it usually solves the normal equation derived from an observation equation. The resistivity inversion method using Householder's transformation solves the observation equation directly, so that it is numerically more stable than the conventional method. A theoretical, ill-conditioned Schlumberger sounding data was chosen to test the inversion scheme with Householder's transformation.