四参数二元McKay型伽马分布的几个性质
PDF下载 (431)李国安 1,李建峰 1,李穆真 2.四参数二元McKay型伽马分布的几个性质[J].宁波大学学报(理工版),2018,31(1):62-68.DOI:
LI Guo-an 1,LI Jian-feng 1,LI Mu-zhen 2.Some properties of the four-parameter McKay’s bivariate gamma distribution[J].Journal of Ningbo University(Natural Science & Engineering Edition),2018,31(1):62-68.DOI:
| Title: | Some properties of the four-parameter McKay’s bivariate gamma distribution |
| 作者: | 李国安 1, 李建峰 1, 李穆真 2 |
| Author(s): | LI Guo-an 1, LI Jian-feng 1, LI Mu-zhen 2 |
| 关键词: | 四参数二元McKay型伽马分布; 相关性质; 渐近性质; 二步拟合; the four-parameter McKay’s bivariate gamma distribution |
| Keywords: | correlation property; asymptotical property; fitting data by way of two steps |
| 分类号: | O211.4 |
| 文献标识码: | A |
| 摘要: | 讨论了四参数二元McKay型伽马分布的几个性质. 通过变量替换技术, 导出了四参数二元McKay型伽马分布的特征及模拟性质; 通过已知可识最小值及差值的分布密度, 所有参数皆可识别, 得到识别性质; 通过计算X, Y之间的相关系数, 证明了相关系数落在(0,1)内, 得到相关性质, 证明了X的尾部关于Y的尾部渐近不相关以及Y的尾部关于X的尾部渐近正相关, 得到渐近性质; 最后, 得到了四参数二元McKay型伽马分布参数的矩估计及最大似然估计. 从应用层面考虑, 采用二步拟合, 先导出了位置参数的最大似然估计, 进而导出尺度参数和形态参数的矩估计, 得到估计性质. 利用已有文献中的实例数据, 给出了基于四参数二元McKay型伽马分布的地震灾害评估参数的估计值. |
| Abstract: | Some properties of the four-parameter Mckay’s bivariate gamma distribution are considered in this paper. By using the technique of variate substitution, the characterization of the McKay’s bivariate gamma distribution are derived, from which an efficient simulation of the Mckay’s bivariate gamma distribution is obtained. Also if the joint distribution of identified minimum and differential value is known, then all of parameters are identified; and correlation coefficient of X and Y is calculated. It is proven that the value of correlation coefficient is between zero and one; it also shows that the tail of X with respect to the tail of Y is asymptotically independent, while the tail of Y with respect to the tail of X is asymptotically positive related. The moment estimators and maximum likelihood estimators of parameters of the four-parameter Mckay’s bivariate gamma distribution are obtained. Finally, taking into account the application, fitting data by way of two steps, the moment estimator of one parameter, the maximum likelihood estimator of another parameter, as well as the modified moment estimators of the other two parameter of the four-parameter Mckay’s bivariate gamma distribution are derived based on integrating the method of moment estimator with the method of maximum likelihood estimator. In the end, the results of fitting to the earthquake data from the reference are provided. |
| 参考文献 /References: | [1] Pasari S, Dikshit O. Three-parameter generalized exponential distribution in earthquake recurrence interval estimation[J]. Nat Hazards, 2014, 73:639-656. [2] Pasari S, Dikshit O. Impact of three-parameter Weibull models in probabilistic assessment of earthquake hazards[J]. Pure and Applied Geophysics, 2014, 171:1251-1281. [3] Pasari S, Dikshit O. Distribution of earthquake interevent times in northeast India and adjoining regions[J]. Pure and Applied Geophysics, 2015, 172:2533-2544. [4] 李国安, 李建峰. 指数分布抽样基本定理及在三参数一般指数分布参数估计中的应用[J].数学的实践与认识, 2017, 47(3):165-169. [5] Hristopulos D T, Mouslopoulou V. Strength statistics and the distribution of earthquake interevent times[J]. Physica A, 2013, 393:485-495. [6] Bacchi B, Bacciu G, Kottegoda N T. Bivariate exponential model applied to intensities and durations of extreme rainfall[J]. Journal of Hydrology, 1994, 155(1/2):225-236. [7] Dargahi-Noubary G R, Razzaghi M. Earth-quake hazard assessment based on bivariate exponential distributions[J]. Reliability Engineering & System Safety, 1994,44(2):153-166. [8] McKay A T. Sampling from batches[J]. Journal of the Royal Statistical Society, 1934(S):207-216. [9] Gupta A K, Nadarajah S. Sums, products and ratios for McKay’s bivariate gamma distribution[J]. Mathematical and Computer Modelling, 2006, 43:185-193. [10] Nadarajah S, Gupta A K. Some bivariate gamma distributions[J]. Applied Mathematics Letters, 2006, 19:767-774. [11] 李国安, 李建峰. 例说基于可识最小值之识别性与参数估计及特征的关系[J]. 大学数学, 2016, 32(4):20-29. [12] Yue S, Ouarda T B M J, Bobée B. A review of bivariate gamma distributions for hydrological application[J]. Journal of Hydrology, 2001, 246(1):1-18. [13] Basu A P, Ghosh J K. Identifiability of the multinorma and other distributions under competing risks model[J]. Journal of Multivariate Analysis, 1978, 8(3):413-429. [14] Domma F. Some properties of the bivariate Burr type III distribution[J]. Journal of Theoretical and Applied Statistics, 2010, 44(2):203-215. [15] 李国安. 二元Marshall~Olkin型指数分布的独立性和不相关性[J]. 大学数学, 2013, 29(4):91-93. [16] Lin G D, Lai C D, Govindaraju K. Correlation structure of the Marshall-Olkin bivariate exponential distribution[J]. Statistical Methodology, 2016, 29:1-9. [17] 李国安, 李建峰. 二元Lawrance-Lewis型指数分布的识别性及渐近不独立性[J]. 高等数学研究, 2016, 19(6):44-47. [18] Kotz S, Navarro J, Ruiz J M. Characterizations of Arnold and Strauss’ and related bivariate exponential models[J]. Journal of Multivariate Analysis, 2007, 98(7):1494-1507. [19] Yu Y M. Efficient simulation of a bivariate exponential conditionals distribution[J]. Computational Statistics & Data Analysis, 2008, 52:2273-2276. [20] Nadarajah S, Kotz S. Reliability models based on bivariate exponential distributions[J]. Probabilistic Engineering Mechanics, 2006, 21:338-351. [21] Gupta R C. Reliability studies of bivariate distribution with exponential conditionals[J]. Mathematical and Computer Modelling, 2008, 47:1009-1018. [22] Vinh A, Griffiths W E, Chotikapanich D. Bivariate income distributions for assessing inequality and poverty under dependent samples[J]. Economic Modelling, 2010, 27:1473-1483. [23] 董永权, 王占民. 关于相关系数P的几点注释[J]. 大学数学, 2008, 24(2):182-186. [24] Marshall A W, Olkin I. A new method for adding a parameter to a family of distributions with application to the exponential and Weibull families[J]. Biometrika, 1997, 84:641-652. [25] Dixit U J, Nasiri P N. Estimation of parameters of a right truncated exponential distribution[J]. Statistical Papers, 2008, 49:225-236. [26] 李国安, 黄林涛, 李建峰. 最大似然估计法的探究式教学[J]. 大学数学, 2013, 29(3):144-146. |
| 备注/Memo: | 收稿日期: 2016－09－27. 宁波大学学报（理工版）网址: http://journallg.nbu.edu.cn/基金项目: 宁波大学学科项目（XKL14D2037）.第一作者: 李国安（1964－）, 男, 浙江慈溪人, 副教授, 主要研究方向: 数理统计与土地估价. E-mail: liguoan@nbu.edu.cn 宁波大学学报（理工版）网址：http://journallg.nbu.edu.cn/ |
