• The Korean Journal of Applied Statistics
• /
• v.33 no.1
• /
• pp.107-114
• /
• 2020

Extreme value distributions have often been used for the analysis (e.g., prediction of return level) of data which are observed from natural disaster. By the extreme value theory, the block maxima asymptotically follow the generalized extreme value distribution as sample size increases; however, this may not hold in a small sample case. For solving this problem, this paper proposes the use of a log-logistic (LLG) distribution whose validity is evaluated through goodness-of-fit test and model selection. The proposed method is illustrated with data from annual maximum earthquake magnitudes of China. Here, we present the predicted return level and confidence interval according to each return period using LLG distribution.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.1 no.4
• /
• pp.59-68
• /
• 1997

In analysis of underground structures, the effects of artificial boundary conditions are considered as one of the major reasons for differences from experimental results. These phenomena can be overcome by using the boundary elements which satisfy the multi-layered half space conditions. The fundamental solutions of multi-layered half-space for boundary element method is formulated satisfying the transmission and reflection of waves at each layer interface and radiation conditions at bottom layer. The governing equations can be obtained from the displacements at each layer which are expressed in terms of harmonic functions. All types of waves can be included using the complete response from semi-infinite integrals with respect to horizontal wavenumbers using expansion of Fourier series and Hankel transformation. Two dimensional Green's functions are derived from cylindrical Navier equations and potentials performing infinite integration in y-direction. In this case, it is effective to transform into two dimensional problem using semi-analytical integration and sinusoidal Bessel function. Some verifications are given to show the accuracy and efficiency of the developed method, and numerical examples to demonstrate the dynamic behavior of underground with various properties.

• The Korean Journal of Applied Statistics
• /
• v.24 no.5
• /
• pp.833-845
• /
• 2011

This paper conducts a statistical analysis of extreme values for both daily log-returns and daily negative log-returns, which are computed using a collection of KOSPI data from January 3, 1998 to August 31, 2011. The Poisson-GPD model is used as a statistical analysis model for extreme values and the maximum likelihood method is applied for the estimation of parameters and extreme quantiles. To the Poisson-GPD model is also added the Bayesian method that assumes the usual noninformative prior distribution for the parameters, where the Markov chain Monte Carlo method is applied for the estimation of parameters and extreme quantiles. According to this analysis, both the maximum likelihood method and the Bayesian method form the same conclusion that the distribution of the log-returns has a shorter right tail than the normal distribution, but that the distribution of the negative log-returns has a heavier right tail than the normal distribution. An advantage of using the Bayesian method in extreme value analysis is that there is nothing to worry about the classical asymptotic properties of the maximum likelihood estimators even when the regularity conditions are not satisfied, and that in prediction it is effective to reflect the uncertainties from both the parameters and a future observation.

• Korean Journal of Plant Resources
• /
• v.30 no.5
• /
• pp.481-488
• /
• 2017

Under circumstances of rapid disappearing of traditional ethnobotanical knowledge, traditional ethnobotanical knowledge surveys are the major step in documenting useful species with a conservation priority. In the ethnobotanical research, the relevance to the survey intensity, ethnobotanical information and plant species richness is the most important research theme. We made up TEK database in south Korea using metadata which had been published by the Korea National Arboretum. We calculated species richness using species richness estimator like ACE, Chao1, Chao2, ICE, Jack 1, Jack 2, and Bootstrap. Species accumulation curves showed each province sampling efforts appeared to be wide range of variance so that Gangwon province need more sampling efforts, and Chungnam province approached a horizontal asymptote earlier. We found heterogeneous patterns in the rarefaction curves of TEK species between gender for each categories of use (medicinal, food and handicrafts). Comparing with regional floral diversities, it was predicted that more diverse species would be found in some provinces by carrying out additional survey.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.5
• /
• pp.515-521
• /
• 2015

The SymMerge algorithm is an efficient merging algorithm for input sequences u and v of sizes $\left|u \right|=m$ and $\left|v \right|=n$, $m{\leq}n$. We consider complexity analysis for SymMerge algorithm regarding to the required number of comparisons. The focus of the previous complexity analysis was on finding the values of upper bounds, i.e. showing the asymptotical optimality. In this paper, in a different way from the previous complexity analysis, we show that the overall required number of comparisons for two representative special cases "symmetric case" and "maximum spanning case" can be calculated exactly i.e. the least upper bounds regarding to the required number of comparisons are calculated. Symmerge requires exactly $m\;log\frac{n}{m}+4m-logm-3$ comparisons for symmetric case of sizes $m=2^k,\;n=2^l,l{\geq}k$ of input sequences and exactly $\frac{1}{2}m^2+(m+1)logm-\frac{3}{2}m+2$ comparisons for maximum spanning case of sizes $m=2^k,n=2^m-m$ of input sequences. Additionally we show that the complexity of the Symmerge algorithm regarding to the overall required number of comparisons for these special cases can be defined by recurrence relations.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.5A
• /
• pp.511-520
• /
• 2011

In this paper, we introduce a multiuser (MU) scheduling method for multiuser amplify-and-forward relay systems, which selects both the transmission mode, i.e., either one- or two-hop transmission, and the desired user via two steps. A closed-form expression for the average achievable rate of the proposed scheduling is derived under two transmission modes with MU scheduling, and its asymptotic solution is also analyzed in the limit of large number of mobile stations. Based on the analysis, we perform our two-step scheduling algorithm: the transmission mode selection followed by the user selection that needs partial feedback for instantaneous signal-to-noise ratios (SNRs) to the base station. We also analyze the average SNR condition such that the MU diversity gain is fully exploited. In addition, it is examined how to further reduce a quantity of feedback under certain conditions. The proposed algorithm shows the comparable achievable rate to that of the optimal one using full feedback information, while its required feedback overhead is reduced below half of the optimal one.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.3
• /
• pp.96-106
• /
• 2009

Extending the scale of the distributed processing in a single Ethernet cluster is physically restricted by maximum ports per switch. This paper presents an implementation of MPI-based multicluster consisting of multiple Ethernet switches for extending the scale of distributed processing, and a asymptotical analysis for communication overhead through execution-time analysis model. To determine an optimum task partitioning, we analyzed the processing time for various partitioning schemes, and AAP(accumulator array partitioning) scheme was finally chosen to minimize the overall communication overhead. The scope of data partitioned in AAP was modified to fit for incremented nodes, and suitable load balancing algorithm was implemented. We tried to alleviate the communication overhead through exploiting the pipelined broadcast and flat-tree based result gathering, and overlapping of the communication and the computation time. We used the linear pipeline broadcast to reduce the communication overhead in intercluster which is interconnected by a single link. Experimental results shows nearly linear speedup by the proposed parallel distributed GHT implemented on MPI-based Ethernet multicluster with four 100Mbps Ethernet switches and up to 128 nodes of Pentium PC.

• Korean journal of applied entomology
• /
• v.40 no.4
• /
• pp.301-307
• /
• 2001

The pine needle gall midge (PNGM), Thecodiplosis japonensis Uchida et Inouye, is a serious pest in pine forests in Korea. Since the first report of PNGM infestation in Korea in 1929, the infestation area has been expanded gradually. In 1975 the distribution of PNGM and its parasitoids were surveyed throughout several infested areas in Korea. Annual survey has been made for the rates of gall formation by PNGM and parasitism by its parasitoids at the monitoring sites including newly infested area in 1975 since 1980. These data were used to examine the relationship between PNGM and its parasitoids for newly invaded areas. The gall forming rate of PNGM and the parasitism by the parasitoid were 34.8 and 1.9% in 1975, respectively, while the gall forming rate and the parasitism were 20.7 and 8.9% in 1985, respectively. The relationships of densities between PNGM and its parsitoid were weak in the early stage of dispersion, but the density of parasitoid was with an asymptotic increase along with PNGMs density increase during the observation period for 10 years.

• Journal of the Korean Geotechnical Society
• /
• v.31 no.7
• /
• pp.13-28
• /
• 2015

This study extended the Lee et al.'s (2015a) solution which improved the existing analytical solution for prediction of the residual pore water pressure into progressive wave and flow coexisting field. At this time, the variation of incident wave period and wave length should be incorporated to Lee et al.'s (2015a) analytical solution, which does not consider flow. For the case of infinite thickness, the new analytical solution using Fourier series was compared to the analytical solution using Laplace transformation proposed by Jeng and Seymour (2007). It was verified that the new solution was identical to the Jeng and Seymour's solution. After verification of the new analytical solution, the residual pore water pressure head was examined closely under various given values of flow velocity's magnitude, direction, incident wave's period and seabed thickness. In each proposed analytical solution, asymptotic approach to shallow depth with the changes in the soil thickness within finite soil thickness was found possible, but not to infinite depth. It is also identified that there exists a discrepancy case between the results obtained from the finite and the infinite seabed thicknesses even on the same soil depth.

• The Korean Journal of Applied Statistics
• /
• v.10 no.1
• /
• pp.69-84
• /
• 1997

We consider a linear calibration problem, $y_i = $$\alpha + \beta (x_i - x_0) + \epsilon_i$, $i=1, 2, {\cdot}{\cdot},n$ $y_f = \alpha + \beta (x_f - x_0) + \epsilon, $ where we observe $(x_i, y_i)$'s for the controlled calibration experiments and later we make inference about $x_f$ from a new observation $y_f$. The objective of the calibration design problem is to find the optimal design $x = (x_i, \cdots, x_n$ that gives the best estimates for $x_f$. We compare Kim(1989)'s Bayesian design which minimizes the expected value of the posterior variance of $x_f$ and some optimal designs from literature. Kim suggested the Bayesian optimal design based on the analysis of the characteristics of the expected loss function and numerical must be equal to the prior mean and that the sum of squares be as large as possible. The designs to be compared are (1) Buonaccorsi(1986)'s AV optimal design that minimizes the average asymptotic variance of the classical estimators, (2) D-optimal and A-optimal design for the linear regression model that optimize some functions of $M(x) = \sum x_i x_i'$, and (3) Hunter & Lamboy (1981)'s reference design from their paper. In order to compare the designs which are optimal in some sense, we consider two criteria. First, we compare them by the expected posterior variance criterion and secondly, we perform the Monte Carlo simulation to obtain the HPD intervals and compare the lengths of them. If the prior mean of $x_f$ is at the center of the finite design interval, then the Bayesian, AV optimal, D-optimal and A-optimal designs are indentical and they are equally weighted end-point design. However if the prior mean is not at the center, then they are not expected to be identical.In this case, we demonstrate that the almost Bayesian-optimal design was slightly better than the approximate AV optimal design. We also investigate the effects of the prior variance of the parameters and solution for the case when the number of experiments is odd.