应用概率统计

 Office Online
 Journal Online

Current Issue

 2009 Vol.25 Issue.5,Published 2009-09-20
 学术论文
 449 A Location Invariant Pickands-Type Estimator Tao Bao In this paper, the author proposesa location invariant Pickands-type estimator as the extreme index ispositive. Its weak and strong convergence are proved. Asymptoticalrepresentation and strong convergence rate are derived, and theoptimal choice of $k_2$ is found. At last simulation and comparisonof this new kind of Pickands-type estimator with other knownPickands-type estimators are considered by using the adaptivealgorithm. 2009 Vol. 25 (5): 449-460 [Abstract] ( 1645 ) [HTML 0KB] [ PDF 312KB] ( 1376 )
 461 The CUSUM Control Chart for the Autocorrelated Data\\with Measurement Error Liu Xiaohong:Wang Zhaojun As we know, the measurement error oftenexists in practice, and affects the performance of quality controlin some cases. The autoregressive process with the measurement erroris investigated in this paper. For detecting the step shift of theautoregressive process mean with measurement error, a CUSUM controlchart based on the maximum log-likelihood ratio test is obtained.Simulated in-control and out-of-control ARL's are made for variousmeasurement error and autocorrelation coefficients. The simulationresults show that this new CUSUM scheme works well when the processis negatively autocorrelated. 2009 Vol. 25 (5): 461-474 [Abstract] ( 1881 ) [HTML 0KB] [ PDF 220KB] ( 1349 )
 475 Assessing Occupational Exposure Via the Unbalanced\\One-Way Random Models Li Xinmin A unbalanced one-way random model is consideredfor assessing the proportion of workers whose mean exposure exceedthe occupational exposure limit (OEL) based on exposure measurementsto the worker. Hypothesis testing for the relevant parameter ofinterest is proposed when the exposure data are unbalanced. Themethod is based on the generalize inference. A simulation study isconducted to compare it with that of Krishnamoorthy and Guo (2005).Simulation results suggest that the proposed method appears to bebetter, especially in very unbalanced design. 2009 Vol. 25 (5): 475-484 [Abstract] ( 1488 ) [HTML 0KB] [ PDF 203KB] ( 1296 )
 485 New Multivariate Skew t Distributions Generated\\from Skew Pearson VII Distributions Wen Yangjun:Zhu Daoyuan In general, there are a large number of skewelliptically contoured distributions with hard calculated integralforms, while skew normal, skew normal scale mixtures, skew Pearsontype VII and skew Pearson type II distributions possess goodstructures. Skew t distributions belong to skew Pearson type VIIdistributions, therefore, this paper proposes new multivariate skewt distributions based on multivariate skew Pearson type VIIdistributions. Backgrounds and definitions are given, two stochasticrepresentations and their equivalence are derived. 2009 Vol. 25 (5): 485-498 [Abstract] ( 1769 ) [HTML 0KB] [ PDF 924KB] ( 1435 )
 499 The Gerber-Shiu Function for a Sparre Andersen\\Risk Model Fan Qingzhu:Yin Chuancun In this paper, we consider a Sparre Andersen risk model in whichthe claim inter-arrival distribution is a mixture of an Erlang($n$)distribution and an Erlang($m$) distribution. Our purpose is tostudy the Gerber-Shiu function $\phi_\delta(u)$. First, we show that$\phi_\delta(u)$ satisfies a higher-order integro-differentialequation, and then we discuss the roots of the generalizedLundberg's equation. On this basis, we drive the Laplace transformof $\phi_\delta(u)$ and prove that $\phi_\delta(u)$ satisfies arenewal equation. Finally, we consider an example. 2009 Vol. 25 (5): 499-512 [Abstract] ( 1946 ) [HTML 0KB] [ PDF 258KB] ( 1274 )
 513 Admissibility of Linear Estimators in Multivariate\\Linear Models with Respect to an Incomplete\\Ellipsoidal Restriction Wu Jianhong This paper studies the admissibility oflinear estimators in multivariate linear models with respect to anincomplete ellipsoidal restriction$\mbox{tr}(\Theta-\Theta_1)'N(\Theta-\Theta_1)\leq\sigma^2$.Specifically, we study the influence of the matrix $N$ and$\Theta_1$ which is the center of a restricted set to theadmissibility of linear estimators in multivariate linear modelswith respect to the incomplete ellipsoidal restriction$\mbox{tr}(\Theta-\Theta_1)'N(\Theta-\Theta_1)\leq\sigma^2$. Themain results show that the class of admissible linear estimatorswith the restriction$\mbox{tr}(\Theta-\Theta_1)'N(\Theta-\Theta_1)\leq\sigma^2$ is thesame as the one with the restriction$\mbox{tr}(\Theta-\Theta_2)'N(\Theta-\Theta_2)\leq\sigma^2$ for$\Theta_1$ and $\Theta_2$ with certain relationship. 2009 Vol. 25 (5): 513-518 [Abstract] ( 1871 ) [HTML 0KB] [ PDF 172KB] ( 1432 )
 519 A Note on Weighted Invariance Principle Li Linyuan:Chen Ping In this note we generalize Davydov's\ucite{1}weak invariance principle for stationary processes to a weightedpartial sums of long memory infinite moving average processes. Thisnote also contains some bounds on the second moments of incrementsof some weighted partial sum processes of a general long memory timeseries, not necessarily moving average type. These bounds are usefulin proving the tightness in uniform metric of these processes. As aconsequence of continuous mapping theorem, the probability bounds oncertain functions of random variables can be established. 2009 Vol. 25 (5): 519-530 [Abstract] ( 1683 ) [HTML 0KB] [ PDF 212KB] ( 1382 )
 531 Inference in Varying-Coefficient Mixed Models\\by Using Smoothing Spline Lu Yiqiang:Xu Wangli Varying-coefficients mixed model (VCMM) isproposed for longitudinal data and the other correlated data. Thismodel allows flexible functional dependence of the response variableon the covariates by using varying-coefficients linear part topresent the covariates effects, while accounting the within-subjectcorrelation by using random effect. In this article, the coefficientfunctions are estimated by using smoothing spline and restrictedmaximum likelihood is used to estimate the smoothing parameters andthe variance components simultaneously. The performance of theproposed method is evaluated though some simulation studies, whichshow that both the coefficient functions and variance componentscould be estimated well for the VCMMs with all kinds of covariancestructures. 2009 Vol. 25 (5): 531-543 [Abstract] ( 1792 ) [HTML 0KB] [ PDF 324KB] ( 1449 )
 544 Expected discounted Penalty Function\\for a Thinning Risk Model Pan Jie:Wang Guojing In this paper, we consider a risk model inwhich the premium arriving number process is a Poisson process withparameter $\lambda>0$ while the claim number process is the thinningof the pre- mium arriving number process. Under such a model, weobtain the integral equation, the integro- differential equation andthe recursive formula for the expected discounted penalty function.Using the integro-differential equation we get the closed formexpressions for the Laplace transform of the time of ruin and thedeficit at ruin when the premium and the claim sizes areexponentially distributed. 2009 Vol. 25 (5): 544-552 [Abstract] ( 1721 ) [HTML 0KB] [ PDF 198KB] ( 1437 )
 553 The Credibility Model under Balance Loss Function Wen Limin:Lin Xia:Wang Jinglong When the insurance company determine thepremium of corresponding risks, the target premium will be usuallyconsidered. In this article, we derived the credibility estimatorsunder the balanced loss function by means of credibility theory. Wealso estimated the unknown parameters and got the empirical Bayescredibility estimators. Finally, we simulated the results under thenormal case which showed the desired conclusions. 2009 Vol. 25 (5): 553-560 [Abstract] ( 1653 ) [HTML 0KB] [ PDF 219KB] ( 1476 )

 News

 ·
 ·
 ·
 · New and modified homepage for APS comes