• Journal of the Korean Data and Information Science Society
• /
• v.22 no.5
• /
• pp.961-966
• /
• 2011

Circular variables are those that have a period in its range. Their examples include direction of animal migration, and time of drug administration, just to mention a few. Statistical analysis of circular variables is quite different from that of linear variable due to its periodic nature. In this paper, the author proposes new circular regression models using geodesic lines on the surface of the sample space of the response and the predictor variables.

• Journal of The Korean Astronomical Society
• /
• v.29 no.spc1
• /
• pp.287-289
• /
• 1996

IRIS(International Research on the Interior of the sun) is the name of a worldwide network of 6 stations for whole disc Doppler shift measurements. The network has been operating since 1987 and by now a few series of a hundred days long unbroken (by day/night periodicity) data were received. Analysis of these data allowed to receivesome new results which are discussed in the paper.

• Journal of the Korean Data and Information Science Society
• /
• v.15 no.4
• /
• pp.911-920
• /
• 2004

We try receiver operating characteristic(ROC) curves by neural networks of logistic function. The models are shown to arise from model classification for normal (diseased) and abnormal (nondiseased) groups in medical research. A few goodness-of-fit test statistics using normality curves are discussed and the performances using neural networks of logistic function are conducted.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.529-531
• /
• 2001

We present an expectation-maximization algorithm for principal component analysis via orthogonalization. The algorithm finds actual principal components, whereas previously proposed EM algorithms can only find principal subspace. New algorithm is simple and more efficient thant probabilistic PCA specially in noiseless cases. Conventional PCA needs computation of inverse of the covariance matrices, which makes the algorithm prohibitively expensive when the dimensions of data space is large. This EM algorithm is very powerful for high dimensional data when only a few principal components are needed.

• Journal of the Korean Statistical Society
• /
• v.2 no.1
• /
• pp.9-15
• /
• 1973

In the first-order moving average model, we present the exact likelihood equations as function of variance, correlation and parameters of coefficients in the orthogonally transformed model. Existence of maximum likelihood estimates for these unknowns are studied and a computational method is provided. (Because of the limited space Ive do not present the computer program which is written in FORTRAN.) 40 sets of generated data and economic data are used to demonstrate, and few of them are presented in the Appendix. A numerical comparison of MLE with the efficient estimate proposed by Durbin is presented in the particular case.

• Proceedings of the Korean Society of Coastal and Ocean Engineers Conference
• /
• 2000.09a
• /
• pp.176-181
• /
• 2000

This paper briefly outline the approach we are now setting for improving the existing global ocean tidal charts in next few years. There has been notable progress in predicting global ocean tide in mid 90s to improve correction procedures for tidal signals in altimetry with more accurate tidal models (http://podaac.jpl.nasa.gov/) than existing ones. (omitted)

• Journal of Applied Reliability
• /
• v.1 no.2
• /
• pp.139-148
• /
• 2001

We are concerned in the estimation of lifetime distribution of a component having a few observed failure data and we demonstrate that we can use the maximum likelihood estimation utilizing the censored data when the number of components in use is quite large. Weibull distribution is assumed for the lifetime and a simple estimation procedure is introduced. We also propose the method of determining the optimal number of spares required for a specified period without introducing cost functions. The proposed methods are applied to the photon stop and the water-cooled flange which are major components in Pohang Light Source.

• Proceedings of the KIEE Conference
• /
• 1988.07a
• /
• pp.198-202
• /
• 1988

In digital mobile communications received speech data are affected by burst errors as well as random errors. To overcome these errors we propose a bit-selective forward error correction scheme for the speech data which is sub-band coded at 13 kbps and transmitted over a 16 kbps channel. For a few error correcting codes the signal-to-noise ratio of error-corrected speech is obtained and compared through the simulation of mobile communication channels.

• Proceedings of the Korean Reliability Society Conference
• /
• 2000.04a
• /
• pp.201-208
• /
• 2000

We are concerned in the estimation of lifetime distribution of a component having a few observed failure data and we demonstrate that we can use the maximum likelihood estimation utilizing the censored data when the number of components in use is quite large. Weibull distribution is assumed fur the lifetime and a simple estimation procedure is introduced. We also propose the method of determining the optimal number of spares required for a specified period without introducing cost functions. The proposed methods are applied to the photon stop and the water-cooled flange which are major components in Pohang Light Source.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.05a
• /
• pp.29-33
• /
• 2001

We propose fuzzy least-squares regression analysis by few error term data and test the slop by fuzzy hypotheses membership function for fuzzy number data with agreement index. Finding the agreement index by area for fuzzy hypotheses membership function and membership function of confidence interval, we obtain the results to acceptance or reject for the test of fuzzy hypotheses.