A Maximum-Entropy Compound Distribution Model for Extreme Wave Heights of Typhoon-Affected Sea Areas

A new compound distribution model for extreme wave heights of typhoon-affected sea areas is proposed on the basis of the maximum-entropy principle. The new model is formed by nesting a discrete distribution in a continuous one, having eight parameters which can be determined in terms of observed data of typhoon occurrence-frequency and extreme wave heights by numerically solving two sets of equations derived in this paper. The model is examined by using it to predict the N-year return-period wave height at two hydrology stations in the Yellow Sea, and the predicted results are compared with those predicted by use of some other compound distribution models. Examinations and comparisons show that the model has some advantages for predicting the N-year return-period wave height in typhoon-affected sea areas.

A COMPOUND POISSON MODEL FOR LEARNING DISCRETE BAYESIAN NETWORKS

We introduce here the concept of Bayesian networks, in compound Poisson model, which provides a graphical modeling framework that encodes the joint probability distribution for a set of random variables within a directed acyclic graph. We suggest an approach proposal which offers a new mixed implicit estimator. We show that the implicit approach applied in compound Poisson model is very attractive for its ability to understand data and does not require any prior information. A comparative study between learned estimates given by implicit and by standard Bayesian approaches is established. Under some conditions and based on minimal squared error calculations, we show that the mixed implicit estimator is better than the standard Bayesian and the maximum likelihood estimators. We illustrate our approach by considering a simulation study in the context of mobile communication networks.

Joint Return Value Estimation of Significant Wave Heights and Wind Speeds with Bivariate Copulas

The joint design criteria of significant wave heights and wind speeds are quite important for the structural reliability of fixed offshore platforms.However,the design method that regards different ocean environmental variables as independent is conservative.In the present study,we introduce a bivariate sample consisting of the maximum wave heights and concomitant wind speeds of the threshold by using the peak-over-threshold and declustering methods.After selecting the appropriate bivariate copulas and univariate distributions and blocking the sample into years,the bivariate compound distribution of annual extreme wave heights and concomitant wind speeds is constructed.Two joint design criteria,namely,the joint probability density method and the conditional probability method,are applied to obtain the joint return values of significant wave heights and wind speeds.Results show that(28.5±0.5)m s^(-1)is the frequently obtained wind speed based on the Atlantic dataset,and these joint design values are more appropriate than those calculated by univariate analysis in the fatigue design.

二项-二维对数正态分布及其在极端海况预测中的应用

Extreme value analysis is an indispensable method to predict the probability of marine disasters and calculate the design conditions of marine engineering.The rationality of extreme value analysis can be easily affected by the lack of sample data.The peaks over threshold(POT)method and compound extreme value distribution(CEVD)theory are effective methods to expand samples,but they still rely on long-term sea state data.To construct a probabilistic model using shortterm sea state data instead of the traditional annual maximum series(AMS),the binomial-bivariate log-normal CEVD(BBLCED)model is established in this thesis.The model not only considers the frequency of the extreme sea state,but it also reflects the correlation between different sea state elements(wave height and wave period)and reduces the requirement for the length of the data series.The model is applied to the calculation of design wave elements in a certain area of the Yellow Sea.The results indicate that the BBLCED model has good stability and fitting effect,which is close to the probability prediction results obtained from the long-term data,and reasonably reflects the probability distribution characteristics of the extreme sea state.The model can provide a reliable basis for coastal engineering design under the condition of a lack of marine data.Hence,it is suitable for extreme value prediction and calculation in the field of disaster prevention and reduction.

Disaster prevention design criteria for the estuarine cities:New Orleans and Shanghai The lesson from Hurricane Katrina

The accurate prediction of the typhoon （hurricane） induced extreme sea environments is very important for the coastal structure design in areas influenced by typhoon （hurricane）. In 2005 Hurricane Katrina brought a severe catastrophe in New Orleans by combined effects of hurricane induced extreme sea environments and upper flood of the Mississippi River. Like the New Orleans City, Shanghai is located at the estuarine area of the Changjiang River and the combined effect of typhoon induced extreme sea en- vironments, flood peak runoff from the Changjiang River coupled with the spring tide is the dominate factor for disaster prevention design criteria. The Poisson-nested logistic trivariate compound extreme value distribution （PNLTCEYD） is a new type of joint probability model which is proposed by compounding a discrete distribution （typhoon occurring frequency） into a continuous multivariate joint distribution （ typhoon induced extreme events）. The new model gives more reasonable predicted results for New Orleans and Shanghai disaster prevention design criteria.

Prediction of Extreme Significant Wave Height from Daily Maxima

For prediction of the extreme significant wave height in the ocean areas where long term wave data are not available, the empirical method of extrapolating short term data (1 similar to3 years) is used in design practice. In this paper two methods are proposed to predict extreme significant wave height based on short-term daily maxima. According to the daa recorded by the Oceanographic Station of Liaodong Bay at the Bohai Sea, it is supposed that daily maximum wave heights are statistically independent. The data show that daily maximum wave heights obey log-normal distribution, and that the numbers of daily maxima vary from year to year, obeying binomial distribution. Based on these statistical characteristics, the binomial-log-normal compound extremum distribution is derived for prediction of extreme significant wave heights (50 similar to 100 years). For examination of its accuracy and validity, the prediction of extreme wave heights is based on 12 years' data at this station, and based on each 3 years' data respectively. The results show that with consideration of confidence intervals, the predicted wave heights based on 3 years' data are very close to those based on 12 years' data. The observed data in some ocean areas in the Atlantic Ocean and the North Sea show it is not correct to assume that daily maximum wave heights are statistically independent; they are subject to Markov chain condition, obeying log-normal distribution. In this paper an analytical method is derived to predict extreme wave heights in these cases. A comparison of the computations shows that the difference between the extreme wave heights based on the assumption that daily maxima are statistically independent and that they are subject to Markov Chain condition is smaller than 10%.

A New Method for Determining Threshold in Using PGCEVD to Calculate Return Values of Typhoon Wave Height

In using the PGCEVD （Poisson-Gumbel Compound Extreme Value Distribution） model to calculate return values of typhoon wave height, the quantitative selection of the threshold has blocked its application. By analyzing the principle of the threshold selection of PGCEVD model and in combination of the change point statistical methods, this paper proposes a new method for quantitative calculation of the threshold in PGCEVD model. Eleven samples from five engineering points in several coastal waters of Guangdong and Hainan, China, are calculated and analyzed by using PGCEVD model and the traditional Pearson type III distribution （P-III） model, respectively. By comparing the results of the two models, it is shown that the new method of selecting the optimal threshold is feasible. PGCEVD model has more stable results than that of P-III model and can be used for the return wave height in every direction.

Design Code Calibration of Offshore Platform Against Typhoon/Hurricane Attacks

Hurricanes Katrina and Rita resulted in the largest number of platforms destroyed and damaged in the history of Gulf of Mexico operations. With the trend of global warming, sea level rising and the frequency and intensity of typhoon increase. How to determine a reasonable deck elevation against the largest hurricane waves has become a key issue in offshore platforms design and construction for the unification of economy and safety. In this paper, the multivariate compound extreme value distribution （MCEVD） model is used to predict the deck elevation with different combination of tide, surge height, and crest height. Compared with practice recommended by American Petroleum Institute （API）, the prediction by MCEVD has probabilistic meaning and universality.

The application of near-infrared spectra micro-image in the imaging analysis of biology samples

In this research,suitable imaging methods were used for acquiring single compound images ofbiology samples of chicken pectorales tissue section,tobacco dry leaf,fresh leaf and plantglandular hair,respectively.The adverse effects caused by the high water content and thethermal effect of near infrared(NIR)light were effectively solved during the experiment procedures and the data procesing.PCA algorithm was applied to the NIR micro-image of chickenpectorales tissue.Comparing the loading vector of PC3 with the NIR spectrum of dry albumen,the information of PC3 was confimmed to be provided mainly by protein,i.e.,the 3rd score imagerepresents the distribution trend of protein mainly.PCA algorithm was applied to the NIR micro-image of tobacco dry leaf.The information of PC2 was confimed to be provided by carbohydrateindluding starch mainly.Compared to the 2nd score image of tobacco dry leaf,the comparedcorelation image with the reference spectrum of starch had the same distribution trend as the 2nd score image.The comparative correla tion images with the reference spectra of protein,glucose,fructose and the total plant alkaloid were acquired to confirm the distribution trend ofthese compounds in tobacco dry leaf respectively.Comparative correlation images of fresh leafwith the reference spectra of protein,starch,fructose,ghucose and water were acquired to confim the distribution trend of these compounds in fresh leaf.Chemimap imaging of plant glandularhair was acquired to show the tubular structure clearly.

On the Distribution of the Minimum or Maximum of a Random Number of i.i.d. Lifetime Random Variables

Statisticians are usually concerned with the proposition of new distributions. In this paper we point out that a unified and concise derivation procedure of the distribution of the minimum or maximum of a random number N of indepen-dent and identically distributed continuous random variables Yi,{i = 1,2,…,N} is obtained if one compounds the probability generating function of N with the survival or the distribution func-tion of Yi. Expressions are then derived in closed form for the density, hazard and quantile func-tions of the minimum or maximum. The methodology is illustrated with examples of the distributions proposed by Adamidis and Loukas (1998), Kus (2007), Tahmasbi and Rezaei (2008), Barreto-Souza and Cribari-Neto (2009), Cancho, Louzada, and Barriga (2011) and Louzada, Roman and Cancho (2011).

Recursive equations for compound distribution with the severity distribution of the mixed type

In this paper, recursive equations are obtained for compound distribution with the number of claims belonging to (a, b)-family and the severity distribution of the mixed type. Numerical methods to solve these equations are presented, and some numerical results are given.

On the Distributions of Two Classes of Multiple Dependent Aggregate Claims

In this paper we examine two classes of correlated aggregate claims distributions, with univariate claim counts and multivariate claim sizes. Firstly, we extend the results of Hesselager [ASTIN Bulletin, 24： 19-32（1994）] and Wang ＆ Sobrero＇s [ASTIN Bulletin, 24：161-166 （1994）] concerning recursions for compound distributions to a multivariate situation where each claim event generates a random vector. Then we give a multivariate continuous version of recursive algorithm for calculating a family of compound distribution. Especially, to some extent, we obtain a continuous version of the corresponding results in Sundt [ASTIN Bulletin, 29：29-45 （1999）] and Ambagaspitiya [Insurance： Mathematics and Economics, 24：301-308 （1999）]. Finally, we give an example and show how to use the algorithm for aggregate claim distribution of first class to compute recursively the compound distribution.

STUDY ON THE TISSUE DISTRIBUTION OF COMPOUNDS AND METABOLITES OF THE MONGOLIAN MEDICINE DIGEDA-4 DECOCTION IN ACUTE LIVER INJURY INDUCED BY D-GaIN

The objective of this study was to study the distribution characteristics of Mongolian drug Digeda-4 decoction in rats with acute liver injury.The Mongolian drug Digeda-4decoction was administered intragastric in rats with acute liver injury induced by D-GalN.The removal of the Liver,spleen,lung,kidney and heart,10%tissue homogenate

Risk estimation based on mixed normal distribution model for diabetes-related hospitalization claims

Medical insurance service, the important part of national healthcare supporting system with a history dating back more than 100 years ago, remains a global challenge because of its high rates of compensation and difficulty in risk control. When developing the diabetes related, hospitalization insurance, we found that the risk loss of the diabetic inpatients does not follow a symmetrical unimodal distribution： in fact, it is hard to describe its risk loses distribution with a single probability distribution model. Therefore, we put forward a risk measurement method based on a mixed normal distributions model for medical insurance of inpatients with diabetes.

Langevin study of neutron emission in the reactions ^(16)O+^(181)Ta and ^(19)F+^(178)Hf

The pre-scission neutrons measured in the reactions ^16O＋^181Ta and ^19F＋^178Hf are studied via a Langevin equation coupled with a statistical decay model. We find that because of the mass asymmetry of different entrance channels, the spin distributions of compound nuclei would be different, consequently, the measured neutrons in these two reactions would also different. This means that the entrance channel will affect the particle emission in the fission process of hot nuclei.