• Communications for Statistical Applications and Methods
• /
• v.10 no.3
• /
• pp.1037-1046
• /
• 2003

We propose an equivariant and robust estimator in multivariate regression model based on the least trimmed squares (LTS) estimator in univariate regression. We call this estimator as multivariate least trimmed squares (MLTS) estimator. The MLTS estimator considers correlations among response variables and it can be shown that the proposed estimator has the appropriate equivariance properties defined in multivariate regression. The MLTS estimator has high breakdown point as does LTS estimator in univariate case. We develop an algorithm for MLTS estimate. Simulation are performed to compare the efficiencies of MLTS estimate with coordinatewise LTS estimate and a numerical example is given to illustrate the effectiveness of MLTS estimate in multivariate regression.

• Communications for Statistical Applications and Methods
• /
• v.6 no.2
• /
• pp.543-564
• /
• 1999

The principal aim of a sequential phase I clinical trial in which the toxicity reponses of a group of patient(s) determine the dose level of the next patient(s) group is to estimate the maximal tolerated dose(MTD) of a new drug, In this paper we compared with a simulation study the performance of the MTD estimates that are determined by a stopping rule in a design and also those that are determined by analyzing the data after a clinical trial is terminated. To the latter belong the mean median mode and maximum likelihood estimates. For the Standard Methods the stopping rule MTD is quite inefficient but the median MTD has a best efficiency and is robust with respect to the three different toxicity curves. The problem of non-convergence of MLE MTD is severe. A more improved MTD estimate is produced by combining the advantages of the various MTD estimates and its efficiency is better than the single median MTD estimate especially for the toxicity curve of an unlucky choice of dose levels. The simulation results suggest that simple types of phase I designs can be combined with relatively standard analytic techniques to provide a more efficient MTD estimate.

• Journal of Advanced Marine Engineering and Technology
• /
• v.36 no.7
• /
• pp.919-926
• /
• 2012

In order to control a LTI(Linear Time Invariant) system subjected to system noise and measurement noise, first of all, it is necessary to estimate the state of system with reliability. Kalman filtering technique has been widely used to estimate the state of the stochastic LTI system with stationary noise characteristics because of its estimation ability versus algorithm simplicity. However, it often fails to estimate the state of the LTI system of which system parameter uncertainty exists partly and/or input uncertainty exists. In this paper, a new estimation technique based on Kalman filter is suggested for stochastic LTI system under parameter uncertainty and/or input uncertainty. A fuzzy estimation algorithm against uncertainties is introduced so as to compensate the state estimate filtered by Kalman filter. In order to verify the state estimation performance of the suggested technique, several simulations are accomplished.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10b
• /
• pp.158-163
• /
• 1992

This paper describes the estimation of the solid friction in mechanical systems by using the extended Kalman filtering techniques. We proposed two stochastic model for the estimation. The one is the 'parametric model' which represents the friction characteristics by an exponential function with unknown parameters. The other is the 'blind model' which does not assume an explicit model but regard the effect of the friction as an unknown input to a known dynamic system. For both models, we give estimation algorithms to generate the filtered estimate and the smoothed estimate with a fixed lag. The filtered estimate can be generated on-line for compensating the solid friction in mechanical systems. Although on-line applications are impossible, the smoothed estimate is more accurate and can be used to grasp precise friction characteristics. Simulation and experimental results arc presented to show the effectiveness of the proposed techniques.

• Communications for Statistical Applications and Methods
• /
• v.20 no.4
• /
• pp.321-327
• /
• 2013

Bayesian and empirical Bayesian methods have become quite popular in the theory and practice of statistics. However, the objective is to often produce an ensemble of parameter estimates as well as to produce the histogram of the estimates. For example, in insurance pricing, the accurate point estimates of risk for each group is necessary and also proper dispersion estimation should be considered. Well-known Bayes estimates (which is the posterior means under quadratic loss) are underdispersed as an estimate of the histogram of parameters. The adjustment of Bayes estimates to correct this problem is known as constrained Bayes estimators, which are matching the first two empirical moments. In this paper, we propose a way to apply the constrained Bayes estimators in insurance pricing, which is required to estimate accurately both location and dispersion. Also, the benefit of the constrained Bayes estimates will be discussed by analyzing real insurance accident data.

• Speech Sciences
• /
• v.11 no.2
• /
• pp.237-245
• /
• 2004

In general, Gaussian mixture model(GMM) is used to estimate the speaker model for speaker identification. The parameter estimates of the GMM are obtained by using the expectation-maximization (EM) algorithm for the maximum likelihood(ML) estimation. However, if the number of mixtures isn't defined well in the GMM, those parameters are obtained inappropriately. The problem to find the number of components is significant to estimate the optimal parameter in mixture model. In this paper, to estimate the optimal number of mixtures, we propose the method that starts from the sufficient mixtures, after, the number is reduced by investigating the mutual information between mixtures for GMM. In result, we can estimate the optimal number of mixtures. The effectiveness of the proposed method is shown by the experiment using artificial data. Also, we performed the speaker identification applying the proposed method comparing with other approaches.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.40 no.3
• /
• pp.35-41
• /
• 1998

Cost estimate system Will be changed from the prime. cost calculation to the historical cost data because the present system has some problems. In this situation, each owner should prepare his own cost estimate system based on the historical cost data. In this study, the standard work items were classified and the criteria of their work amount computation were established for rural improvement projects. And also the historical cost data were collected in all range of rural improvement projects, and the database system, "HICOMS" (HIstorical COst data Management System) was built. In order to test the applicability of the HICOMS, standard work cost and contractor cost were compared. The results by HICOMS showed high significance and it was concluded that the HICOMS could be applicable for the cost estimate of the rural improvement projects. projects.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.4
• /
• pp.153-159
• /
• 2009

In this paper, we deal with the coordinate estimate algorithm of the wired tablet device. In order to got the position of the electric pen on the tablet, electromagnetic induction effect is used. Most electric tablets have used the electromagnetic induction effect. Tracking the position of the pen on a tablet that is directly related with the performance of the device is very important. In this research, a new real time coordinate estimate algorithm is introduced. To estimate the position of the pen, the electromagnetic induced signals og the wired tablet are used.

• Communications for Statistical Applications and Methods
• /
• v.13 no.2
• /
• pp.273-283
• /
• 2006

In bioassay, the response curve is usually assumed monotone increasing, but its exact form is unknown, so it is very difficult to select the proper functional form for the parametric model. Therefore, we should probably use the nonparametric regression model rather than the parametric model unless we have at least the partial information about the true response curve. However, it is well known that the nonparametric regression estimate is not necessarily monotone. Therefore the monotonizing transformation technique is of course required. In this paper, we compare the finite sample properties of the monotone transformation methods which can be applied to the local linear quasi-likelihood response curve estimate.

• Journal of the military operations research society of Korea
• /
• v.12 no.2
• /
• pp.68-81
• /
• 1986

This thesis studies the estimation from interarrival and service time data of the mean number of customers in service at time t for an $M/G/{\infty}$ queue. The assumption is that the parametric form of the service time distribution is unknown and the empirical distribution of twe service time is used in the estimate the mean number of customers in service. In the case in which the customer arrival rate is known the distribution of the estimate is derived and an approximate normal confidence interval procedure is suggested. The use of the nonparametric methods, which are the jackknife and the bootstrap, to estimate variability and construct confidence intervals for the estimate is also studied both analytically and by simulation.