20 September 2008, Volume 24 Issue 5
    

  • Select all
    |
    学术论文
  • Pricing American Contingent Claims with Frictions
    Meng Qingxin, Lao Lanjun, Zhao Xuelei
    CHINESE JOURNAL OF APPLIED PROBABILITY AND STATIST. 2008, 24(5): 449-262.
    Abstract ( ) Download PDF ( ) Knowledge map Save
    The paper addresses the problem of pricing American Contingent Claims (ACCs) under constraints on portfolio choice and a higher interest rate for borrowing than for lending. In this paper, the formulae of the upper hedging price $h_{\text{up}}(K)$ and the lower hedging price $h_{\text{low}}(K)$ of an ACC is derived by introducing a family of auxiliary frictionless financial markets. Furthermore, the
    arbitrage-free interval $[h_{\text{low}}(K),h_{\text{up}}(K)]$ is identified, based on the principle of absence of arbitrage. In the end, for several concrete constraints on portfolio, explicit computations or estimations of the upper hedging price and the lower hedging price are carried out in the case of American call-option.
  • On Optimising the Estimation of Extreme Value Quantiles of a Probability Distribution
    Ouyang Zisheng, Yang Xiangqun,Chen Neiping
    CHINESE JOURNAL OF APPLIED PROBABILITY AND STATIST. 2008, 24(5): 463-474.
    Abstract ( ) Download PDF ( ) Knowledge map Save
    In this paper, a new extreme value quantile estimator is given and its limit properties are discussed. Under the asymptotic second moment principle, recurring to sub-sample bootstrap method, the optimality problem of sample fraction in extreme value quantile estimation is solved, and the limit properties are proved. We prove our sample fraction is optimal under the asymptotic second moment principle, also an adaptive bootstrap procedure is given.
  • 基于最大惩罚似然的高斯混合模型无监督分类研究
    Yu Peng, Tong Xinwei, Feng Jufu
    CHINESE JOURNAL OF APPLIED PROBABILITY AND STATIST. 2008, 24(5): 475-483.
    Abstract ( ) Download PDF ( ) Knowledge map Save
    In this paper we propose an unsupervised classification algorithm which is based on Gaussian mixture models. Thinking that EM algorithm will result in a local optimal resolution of Gaussian mixture models in parameter estimations, we substitute invert Wishart distribution for Jeffery prior. Experiments show that this
    algorithm improves correct rates and decreases time while estimating
    classifications.
  • Efficient Subset for Portfolio with Singular Covariance Matrix
    Jiang Chunfu, Dai Yonglong
    CHINESE JOURNAL OF APPLIED PROBABILITY AND STATIST. 2008, 24(5): 484-492.
    Abstract ( ) Download PDF ( ) Knowledge map Save
    Szeg\"{o} proposed that efficient subset could occur in the optimal portfolio selection problem with singular covariance matrix. An equivalent definition of efficient subset is given based on the analytic solutions of efficient portfolio in this paper. Some
    necessary and sufficient conditions for existing efficient subset in
    the stock market and for determining whether a stock subset is
    efficient one or not are also derived from the equivalent
    definition.
  • On the Number of Claims before Recovery in the Risk Models with Constant Interest Rate
    Liu Li
    CHINESE JOURNAL OF APPLIED PROBABILITY AND STATIST. 2008, 24(5): 493-500.
    Abstract ( ) Download PDF ( ) Knowledge map Save
    In this paper, we study the probabilities of having $n$ claims before the surplus process recovers to non-negative values and after the process has been ruined, and obtain the respective recursive formulas.
  • Inverse Regression Estimation for Regression Coefficient of Multinomial Probit Model
    Ma Jianjun, Xu Xingzhong
    CHINESE JOURNAL OF APPLIED PROBABILITY AND STATIST. 2008, 24(5): 501-512.
    Abstract ( ) Download PDF ( ) Knowledge map Save
    In this paper, we generalize the Multinomial Probit Model and investigate the inverse regression of the Generalized Multinomial Probit Model. A method of inverse regression is proposed for estimating the regression coefficient. Under some assumptions we obtain the asymptotic normality of the estimate. The result of simulation shows that the method performs
    well.
  • The Modification and Optimization of Haezendonck Risk Measure
    Xun Li, Song Lixin, Wang Dehui
    CHINESE JOURNAL OF APPLIED PROBABILITY AND STATIST. 2008, 24(5): 513-521.
    Abstract ( ) Download PDF ( ) Knowledge map Save
    This paper describes the modification of the Haezendonck risk measure. When the risk becomes very high, the lose or the surplus will become much larger, whose function should be a curve steeper than the straight line on the tail. So based on the Haezendonck risk measure the modified Haezendonck risk measure is proposed. And some properties of the modified Haezendonck risk measure are proved, which has certain directive significance to actual life.
  • The Method Based on the Isotonic Regression Estimation for Identifying the Maximum Tolerated Dose
    Wang Xueli, Zhang Zhongzhan, Shi Ningzhong
    CHINESE JOURNAL OF APPLIED PROBABILITY AND STATIST. 2008, 24(5): 522-530.
    Abstract ( ) Download PDF ( ) Knowledge map Save
    One of the primary objectives of a phase I study is to evaluate toxicity at different dose levels and recommend the maximum tolerated dose (MTD). We give an extend up-and-down design method based on the isotonic regression estimation for identifying MTD. Through a lot of simulations, it shows that the design based on isotonic regression for identifying the maximum tolerated dose has improved the accuracy of recommending MTD. It apparently protects patients efficiently and
    prevents patients from excessive trials under the higher toxicity dose level and makes a great significance in phase I clinical trials design.
    • 论文
  • The Estimation of Distribution Function\\and Its Convergent Rate with Interval Censored Data
    Ding Bangjun
    CHINESE JOURNAL OF APPLIED PROBABILITY AND STATIST. 2008, 24(5): 531-540.
    Abstract ( ) Download PDF ( ) Knowledge map Save
    This paper generalizes the result in [11] to the
    k-point uniform distribution. With this result, we then propose a
    procedure to approximate the cumulative distribution functions of
    continuous variables with k-point uniform distribution functions.
    It is also proved that the convergence rate of the distribution
    functions estimated from the interval data is O(n)^-2/9 under
    suitable conditions.

    • 学术论文
  • The Testing in GEE Method without PA Condition
    Ji Yunqi, Zhu Zhongyi
    CHINESE JOURNAL OF APPLIED PROBABILITY AND STATIST. 2008, 24(5): 541-553.
    Abstract ( ) Download PDF ( ) Knowledge map Save
    Marginal regression model and its associated generalized estimating equation (GEE) are becoming increasingly being used in longitudinal studies. Pepe and Anderson (1994) pointed out that there is an important assumption called PA condition behind GEE method. If the assumption is violated and nondiagonal working correlation matrix is used in GEE, the statistical inference may be
    deficient. This paper focused on PA condition's influence on the testing of regression coefficients in GEE method by theoretic and numeric analysis. Due to the violation of the PA condition, the distributions of Wald statistics and Score statistics based on GLS estimators are noncentral $\chi^{2}$ distributions. The efficiency of testing based on the GEE method is largely influenced.
Chinese Journal of Applied Probability and Statistics (Bimonthly)
（started publication in 1985）
Superintendent: Chinese Society of Probability and Statistics
Editor-in-Chief: Chen Mufa
ISSN 1001-4268
CN 31-1256
Mailing code: 4-414
Download More>>
Links More>>
Visited
    Total visitors:
    Visitors of today:
    Now online:
Share:
Copyright © CHINESE JOURNAL OF APPLIED PROBABILITY AND STATISTICS
Tel: 021-54345267　Fax: 021-54345058　E-mail: aps@stat.ecnu.edu.cn
Powered by Beijing Magtech Co. Ltd.