• The Korean Journal of Applied Statistics
• /
• v.30 no.3
• /
• pp.363-376
• /
• 2017

The insurance market is saturated and its growth engine is exhausted; consequently, the insurance industry is now in a low growth period with insurance companies that face a fierce competitive environment. In such a situation, it will be an important issue to find the probability distributions that can explain the flow of insurance claims, which are the basis of the actuarial calculation of the insurance product. Insurance claims are generally known to be well fitted by lognormal distributions or Pareto distributions biased to the left with a thick tail. In recent years, skew normal distributions or skew t distributions have been considered reasonable distributions for describing insurance claims. Cooray and Ananda (2005) proposed a composite lognormal-Pareto distribution that has the advantages of both lognormal and Pareto distributions and they also showed the composite distribution has a higher fitness than single distributions. In this paper, we introduce new composite distributions based on skew normal distributions or skew t distributions and apply them to Danish fire insurance claim data and US indemnity loss data to compare their performance with the other composite distributions and single distributions.

• Journal of KIISE
• /
• v.42 no.7
• /
• pp.901-909
• /
• 2015

In this paper, to build a parallel corpus between Korean and English in Wikipedia. We proposed a method to find similar sentences based on language resources and topic modeling. We first applied language resources(Wiki-dictionary, numbers, and online dictionary in Daum) to match word sequentially. We construct the Wiki-dictionary using titles in Wikipedia. In order to take advantages of the Wikipedia, we used translation probability in the Wiki-dictionary for word matching. In addition, we improved the accuracy of sentence similarity measuring method by using word distribution based on topic modeling. In the experiment, a previous study showed 48.4% of F1-score with only language resources based on linear combination and 51.6% with the topic modeling considering entire word distributions additionally. However, our proposed methods with sequential matching added translation probability to language resources and achieved 9.9% (58.3%) better result than the previous study. When using the proposed sequential matching method of language resources and topic modeling after considering important word distributions, the proposed system achieved 7.5%(59.1%) better than the previous study.

• Proceedings of the National Institute of Ecology of the Republic of Korea
• /
• v.3 no.3
• /
• pp.149-153
• /
• 2022

The degree distribution of the plant-pollinator network was identified by analyzing the data in the ecosystem and reproduced by a model of the growing bipartite mutualistic networks. The degree distribution of pollinator shows power law or stretched exponential distribution, while plant usually shows stretched exponential distribution. In the growth model, the plant and the pollinator are selected with probability P_p and P_A=1-P_p, respectively. The number of incoming links for the plant and the pollinator is l_p and l_A, respectively. The probability that the link of the plant selects the pollinator of the existing network given as $A_{k_i}=k^{{\lambda}_A}_i/{\sum}_i\;k^{{\lambda}_A}_i$, and the probability that the pollinator selects the plant is $P_{k_i}=k^{{\lambda}_p}_i/{\sum}_i\;k^{{\lambda}_p}_i$. When the nonlinear growth index is 𝛌_X=1 (X=A or P), the degree distribution follows a power law, and if 0≤𝛌_X<1, the degree distribution follows a stretched exponential distribution. The cumulative degree distributions of plants and pollinators of 14 empirical plant-pollinators included in Interaction Web Database were calculated. A set of parameters (P_A,P_P,l_A,l_P) that reproduces these cumulative degree distributions and a growth index 𝛌_X (X=A or P) were obtained. We found that animal takes very heterogenous connections, whereas plant takes a more flexible connection network.

• Wind and Structures
• /
• v.35 no.1
• /
• pp.21-34
• /
• 2022

Gust factor is an important parameter for the conversion between peak gust wind and mean wind speed used for the structural design and wind-related hazard mitigation. The gust factor of typhoon wind is observed to show a significant dispersion and some differences with large-scale weather systems, e.g., monsoons and extratropical cyclones. In this study, insitu measurement data captured by 13 meteorological towers during a strong typhoon Morakot are collected to investigate the statistical characteristics, height and wind speed dependency of the gust factor. Onshore off-sea and off-land winds are comparatively studied, respectively to characterize the underlying terrain effects on the gust factor. The theoretical method of peak factor based on Gaussian assumption is then introduced to compare the gust factor profiles observed in this study and given in some building codes and standards. The results show that the probability distributions of gust factor for both off-sea winds and off-land winds can be well described using the generalized extreme value (GEV) distribution model. Compared with the off-land winds, the off-sea gust factors are relatively smaller, and the probability distribution is more leptokurtic with longer tails. With the increase of height, especially for off-sea winds, the probability distributions of gust factor are more peaked and right-tailed. The scatters of gust factor decrease with the mean wind speed and height. AS/NZ's suggestions are nearly parallel with the measured gust factor profiles below 80m, while the fitting curve of off-sea data below 120m is more similar to AIJ, ASCE and EU.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2005.05b
• /
• pp.1053-1057
• /
• 2005

The occurrence probability (OP) distributions of tide levels using harmonic constants of six tidal gauging stations in Korean coastal zone were estimated and analysed in detail. OP analysis using harmonic constants data of Incheon(Youldo), Mokpo, Yeosu, Pusan, Pohang and Sokcho was carried out and compared with the OP using hourly tidal elevation data which were served through the Internet Homepage by the National Ocean Research Institute. The tidal elevation data were divided by the AHHW (ALLW) value referenced to MSL in order to compare the OP patterns in a relative scale. The OP of the tidal elevation calculated using 38 harmonic tidal constituents relatively well agreed with those of hourly observed tidal elevation data. However, the OP results using four harmonic tidal constituents overestimate the occurrence probability at the peak points and underestimate at the tail-regions of the OP. Especially, the OP patterns of the Sokcho and Pohang tidal gauging stations on the East Sea show totally different patterns and the estimation method using four harmonic constants should be modified and application should be strictly limited on the East Sea areas. The OP patterns are considerably well generated in case of the OP generation using the additional two or three dominant tidal constituents,

• Water for future
• /
• v.28 no.2
• /
• pp.145-154
• /
• 1995

The various analyses of the historical rainfall data need to be utilized in a hydraulic engineering project. The probability distributions of the rainfall events according to annual maximum continuous rainfall depths are studied for the hydrologic frequency analysis. The bivariate normal distribution, the bivariate lognormal distribution, and the bivariate gamma distribution are applied to the rainfall events composed of rainfall depths and its durations at Kangnung, Seoul, Incheon, Chupungnyung, Teagu, Jeonju, Kwangju, and Busan. These rainfall events are fitted to the the bivariate normal distribution and the bivariate lognormal distribution, but not fitted to the bivariate gamma distribution. Frequency curves of probability rainfall events are suggested from the probability distribution selected by the goodness-of-fit test.

• Journal of Welding and Joining
• /
• v.23 no.6
• /
• pp.61-66
• /
• 2005

A probabilistic approach for evaluating failure risk is suggested in this paper. Probabilistic fracture analyses were performed for a pressurized pipe of a Cr-Mo steel reflecting variation of material properties at high temperature. A crack was assumed to be located along the weld fusion line. Probability density functions of major variables were determined by statistical analyses of material creep and creep crack growth data measured by the previous experimental studies by authors. Distributions of these variables were implemented in Monte Carlo simulation of this study. As a fracture parameter for characterizing growth of a fusion line crack between two materials with different creep properties, $C_t$ normalized with $C^*$ was employed. And the elapsed time was also normalized with tT, Resultingly, failure probability as a function of operating time was evaluated fur various cases. Conventional deterministic life assessment result was turned out to be conservative compared with that of probabilistic result. Sensitivity analysis for each input variable was conducted to understand the most influencing variable to the analysis results. Internal pressure, creep crack growth coefficient and creep coefficient were more sensitive to failure probability than other variables.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.228-232
• /
• 2017

Extreme hydrological events can cause serious threats to the society. Hence, the selection of probability distributions for extreme rainfall is a fundamental issue. For this reason, this study was focused on understanding possible distributional changes in annual daily maximum rainfalls (AMRs) over time in Sri Lanka using quantile regression. A simplified nine-category distributional-change scheme based on comparing empirical probability density function of two years (i.e. the first year and the last year), was used to determine the distributional changes in AMRs. Daily rainfall series of 13 station over Sri Lanka were analyzed for the period of 1960-2015. 4 distributional change categories were identified for the AMRs. 5 stations showed an upward trend in all the quantiles (i.e. 9 quantiles: from 0.05 to 0.95 with an increment of 0.01 for the AMR) which could give high probability of extreme rainfall. On the other hand, 8 stations showed a downward trend in all the quantiles which could lead to high probability of the low rainfall. Further, we identified a considerable spatial diversity in distributional changes of AMRs over Sri Lanka.

• Journal of Korean Society for Quality Management
• /
• v.48 no.4
• /
• pp.597-608
• /
• 2020

Purpose: The purpose of this paper is to present a probability distribution of the burst pressure of API 5L X65 pipes for the reliability assessment of corroded gas pipelines. Methods: Corrosion is a major cause of weakening the residual strength of the pipe. The mean residual strength on the corrosion defect can be obtained using the burst pressure code. However, in order to obtain the pipe reliability, a probability distribution of the burst pressure should be provided. This study is concerned with estimating the burst pressure distribution using Monte Carlo simulation. A response surface method is employed to represent the distribution parameter as a model of the corrosion defect size. Results: The experimental results suggest that the normal or Weibull distribution should be suitable as the probability distribution of the burst pressure. In particular, it was shown that the probability distribution parameters can be well predicted by using the depth and length of the corrosion defect. Conclusion: Given a corrosion defect on the pipe, its corresponding burst pressure distribution can be provided at instant. Subsequently, a reliability assessment of the pipe is conducted as well.

• Geomechanics and Engineering
• /
• v.34 no.3
• /
• pp.285-301
• /
• 2023

During the construction of E75 highway through Grdelica gorge in Serbia, a major failure occurred in the zone of reinforced rock slope. Excavation was performed in highly anisotropic Paleozoic schist rock formation. The reinforcement consisted of the two rows of micropile wall with pre-stressed anchors. Forces in anchors were monitored with load cells while benchmarks were installed for superficial displacement measurements. The aim of the study is to investigate possible causes of instability considering different probability distributions of the strength of discontinuities and anchor bond strength by applying different optimization techniques for finding the critical failure surface. Even though the deterministic safety factor value is close to unity, the probability of failure is governed by variability of shear strength of anisotropic planes and optimization method used for locating the critical sliding surface. The Cuckoo search technique produces higher failure probabilities compared to the others. Depending on the assigned statistical distribution of input parameters, various performance functions of the factor of safety are obtained. The probability of failure is insensitive to the variation of bond strength. Different sampling techniques should yield similar results considering that the sufficient number of safety factor evaluations is chosen to achieve converged solution.