• Communications for Statistical Applications and Methods
• /
• v.19 no.5
• /
• pp.639-646
• /
• 2012

The exponential continuous time GARCH(p,q) model for financial assets suggested by Haug and Czado (2007) is considered, where the log volatility process is driven by a general L$\acute{e}$vy process and the price process is then obtained by using the same L$\acute{e}$vy process as driving noise. Uniform ergodicity and ${\beta}$-mixing property of the log volatility process is obtained by adopting an extended generator and drift condition.

• Journal of the Korean Data and Information Science Society
• /
• v.16 no.4
• /
• pp.745-758
• /
• 2005

This paper presents a statistical analysis on the position-specific distributions of amino acid residues in transmembrane proteins. A hidden Markov model segments membrane proteins to produce segmented regions of homogeneous statistical property from variable-length amino acids sequences. These segmented residues are analyzed by using chi-square statistic and relative-entropy in order to find position-specific amino acids. This analysis showed that isoleucine and valine concentrated on the center of membrane-spanning regions, tryptophan, tyrosine and positive residues were found frequently near both ends of membrane.

• Communications for Statistical Applications and Methods
• /
• v.2 no.2
• /
• pp.85-93
• /
• 1995

Hannan and Rissanen proposed an innovation regression method of ARMA modeling, which is composed of three stages. Its second-stage is to choose orders of the ARMA model using the BIC, which needs a lot of calculation to estimate several regression models. We are going to present a simple and efficient algorithm for the second stage using a special property of triangular Toeplitz matrices.

• Communications for Statistical Applications and Methods
• /
• v.9 no.2
• /
• pp.411-423
• /
• 2002

In this paper, we propose a split criterion for binary decision trees. The proposed criterion selects the optimal split by measuring the prediction success of the candidate splits at a given node. The criterion is shown to have the property of exclusive preference. Examples are given to demonstrate the properties of the criterion.

• Journal of the Korean Statistical Society
• /
• v.36 no.4
• /
• pp.557-578
• /
• 2007

In Response Surface Methodology (RSM), rotatability is a natural and highly desirable property. For second order general correlated regression model, the concept of robust rotatability was introduced by Das (1997). In this paper a new measure of robust rotatability for second order response surface designs with correlated errors is developed and illustrated with an example. A comparison is made between the newly developed measure with the previously suggested measure by Das (1999).

• Communications for Statistical Applications and Methods
• /
• v.12 no.2
• /
• pp.463-472
• /
• 2005

In this paper, the problem of estimating a p-variate (p $\ge$4) normal mean vector is considered in decision-theoretic set up. Using a simple property of the noncentral chi-square distribution, a sequence of estimators dominating the Lindley type estimator with the cases of unknown covariance matrices has been produced and each improved estimator is better than previous one.

• Communications for Statistical Applications and Methods
• /
• v.11 no.3
• /
• pp.563-574
• /
• 2004

Slope rotatability of response surface designs is a desirable property when we are interested in estimating slopes of response surfaces. In this paper, we derive a simplified measure of slope rotatability from new viewpoints for response surface designs that are frequently used in response surface methodology.

• Journal of the Korean Statistical Society
• /
• v.14 no.1
• /
• pp.29-38
• /
• 1985

Cyclic Factorial Association Scheme (CFAS) for incomplete block designs in a factorial experiment is defined. It is a generalization of EGD/($2^n-1$)-PBIB designs defined by Hinkelmann (1964) or Binary Number Association Scheme (BNAS) named by Paik and Federer (1973). A property of PBIB designs having CFAS is investigated and it is shown that the structural matrix NN' of such designs has a pattern of multi-nested block circulant matrix. The generalized inverse of (rI-NN'/k) is obtained. Generalized Cyclic incomplete block designs for factorial experiments introduced by John (1973) are presented as the examples of CFAS-PBIB designs. Finally, the relationship between CFAS and BNAS in block designs is briefly discussed.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.11 no.1
• /
• pp.73-84
• /
• 1986

A statistical method is proposed which estimates an edge that is one of the basic features in image understanding. The conventional edge detection techniques are performed well for a deterministic singnal, but are not satisfactory for a statistical signal. In this paper, we use the likelihood function which takes account of the statistical property of a signal, and derive the decision function from it. We propose the maximum likelihood edge detection technique which estimates an edge point which maximizes the decision function mentioned above. We apply this technique to statistecal signals which are generated by using the random number generator. Simnulations show that the statistical edge detection technique gives satisfactory results. This technique is extended to the two-dimensional image and edges are found with a good accuracy.

• Journal of Biomedical Engineering Research
• /
• v.17 no.4
• /
• pp.433-442
• /
• 1996

Since one texture property(i.e coarseness, orientation, regularity, granularity) for ultrasound liver ages was not sufficient enough to classify the characteristics of livers, we used multi texture vectors tracted from ultrasound liver images and a statistical classifier. Multi texture vectors are selected among the feature vectors of the normal liver, fat liver and cirrhosis images which have a good separability in those ultrasound liver images. The statistical classifier uses multi texture vectors as input vectors and classifies ultrasound liver images for each multi texture vector by the Bayes decision rule. Then the decision of the liver disease is made by choosing the maximum value from the averages of a posteriori probability for each multi texture vector In our simulation, we obtained higtler correct ratio than that of other methods using single feature vector, for the test set the correct ratio is 94% in the normal liver, 84% in the fat liver and 86% in the cirrhosis liver.