Mortality rate modeling of joint lives and survivor insurance contracts tested by a novel unilateral dependence measure

Recently, nonsymmetric measures of dependence have started to attract attention, and several continuous entropy-like nonsymmetric dependence measures have been proposed. Based on Onicescu's information energy, we have introduced in previous work a nonsymmetric dependence measure between two discrete random variables. In the present paper, we analyze the continuous version of this measure. We deduct that there are important differences when switching from the discrete to the continuous measure. Then we apply this continuous unilateral dependence measure to a real-world challenge: mortality rate modeling for life insurance industry. We consider joint male-female pairs (married, but not necessarily from the same family) belonging to the same policy group, and analyze the unilateral interactions between the male-female mortality data samples for the purpose of stochastic model testing and validation of risk variables in the insurance world.

[1]  R. Andonie,et al.  Energy generalized LVQ with relevance factors , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[2]  Angel Cataron,et al.  An informational energy LVQ approach for feature ranking , 2004, ESANN.

[3]  M. D. Mostafa,et al.  On the problem of estimation for the bivariate lognormal distribution , 1964 .

[4]  Angel Cataron,et al.  kNN estimation of the unilateral dependency measure between random variables , 2014, 2014 IEEE Symposium on Computational Intelligence and Data Mining (CIDM).

[5]  Angel Cataron,et al.  Energy Supervised Relevance Neural Gas for Feature Ranking , 2010, Neural Processing Letters.

[6]  Thomas M. Cover,et al.  Elements of Information Theory (Wiley Series in Telecommunications and Signal Processing) , 2006 .

[7]  Deniz Erdogmus,et al.  Information Theoretic Learning , 2005, Encyclopedia of Artificial Intelligence.

[8]  Angel Cataron,et al.  Informational Energy Kernel for LVQ , 2005, ICANN.

[9]  B. Schweizer,et al.  On Nonparametric Measures of Dependence for Random Variables , 1981 .

[10]  C. B. Bell.  Mutual Information and Maximal Correlation as Measures of Dependence , 1962 .

[11]  John G. van Bosse,et al.  Wiley Series in Telecommunications and Signal Processing , 2006 .

[12]  Dick E. Boekee,et al.  A generalized class of certainty and information measures , 1982, Inf. Sci..

[13]  On Nonsymmetric Nonparametric Measures of Dependence , 2015, 1502.03850.

[14]  Silviu Guiaşu,et al.  Information theory with applications , 1977 .

[15]  P. J. Green,et al.  Density Estimation for Statistics and Data Analysis , 1987 .

[16]  A. Rényi.  On measures of dependence , 1959 .

[17]  Angel Cataron,et al.  Financial data analysis using the informational energy unilateral dependency measure , 2015, 2015 International Joint Conference on Neural Networks (IJCNN).

[18]  Bernard W. Silverman,et al.  Density Estimation for Statistics and Data Analysis , 1987 .