Search

Aiming at the problem of low accuracy of outlier recognition for spatio-temporal data, a framework was constructed according to fusion thought of time dimension and space dimension. Based on the framework, the difference of attribute data between different positions was derived by Archimedean copulas function under the condition of unknown distribution. A method for converting spatial data was established with high value as a core to build rank series. Then, the expectation and variance were determined for hypothesis test. Finally, with the model parameters of window size and scope radius, an approach of outlier recognition for spatio-temporal data was given based-on M-K test. The calculation example and application analysis show that this approach can improve the accuracy of outlier recognition for spatio-temporal data, and has more recognition capability. [ABSTRACT FROM AUTHOR]

The actual distribution and the dependence of financial asset returns are analyzed based on the character of Copula in this paper, a multivariate distribution function which can reflect the actual distribution and the dependence of financial asset returns is developed. An empirical research is done on the performance of the portfolio selection by Monte Carlo method before and after financial crisis in order to research the effect of measuring the actual distribution and dependence on portfolio selection of financial products. [ABSTRACT FROM AUTHOR]

This paper studied the reasonable estimation of factor coefficients when the multi-factor model is used to describe the return rates of financial assets. New methods to estimate factor coefficients were proposed by examining financial implication of factor coefficients and utilizing the Copula theory. Through modeling the joint distribution of each factor and the return rate of the financial asset with the Archimedean Copula function, the new method estimated factor coefficients in two steps: the sign of the coefficient and its occurring probability were first determined by the Kendall's tam the fluctuation scale of the factor coefficient was then calculated as the ratio of the variation of the financial asset's return rate with respect to the change of each factor. Moreover, the above method was further improved by emphasizing the influence of distribution tails on the factor coefficient, which was done by introducing the tail correlation. Empirical tests were carried out with trading data from Chinese stock markets. The practicality and advantage of our new methods were demonstrated by calculating the R-square value, the mean-square deviation of the random error term, and the corresponding tail mean-square deviation of the determined multi-factor models with different factor coefficient estimation methods, as well as their application to the value-at-risk and conditional value-at-risk calculations of the portfolio. Empirical results show that the improved new method is the best, while the new method is obviously better than the traditional linear regression method. [ABSTRACT FROM AUTHOR]

Measuring the credit risk and economic capital is one of the most important aims in risk management for commercial banks. The authors use Johnson transformation to solve the problem of economic capital under non-Gaussian data. By transforming real data to standard normal distribution, the authors overcome the restricts of Monte Carlo simulation in Copula method, which often requires the Gaussian or t distribution. It is convenient to simulate the correlated default time and the economic capital for banks. From the comparative analysis, the Johnson transformation is feasible and reasonable. This paper provides an effective quantitative method in risk management for commercial banks under "Basel Capital Accord" (Basel II) and the algorithm has a certain reference. [ABSTRACT FROM AUTHOR]

Cross-sectional dependence is often encountered in study. The MLE estimators are inefficient when cross section independence is assumed. Therefore, Copula-based log-likelihood functions are introduced to consider the cross section dependence. To avoid the nuisance parameters problem, a method called two-step MLE is proposed. The consistency and the asymptotic normality of the estimator are proved. And the asymptotic covariance matrix is given as well. Monte Carlo experiments show that the two-step MLE improves the efficiency of the estimators significantly. This model is applied to study the characters that affect the cash dividend policies of the listed companies, and it is shown that the sale growth rate becomes more significant relatively. [ABSTRACT FROM AUTHOR]