The Complex-Number Mortality Model (CNMM) based on orthonormal expansion of membership functions Cover Image

The Complex-Number Mortality Model (CNMM) based on orthonormal expansion of membership functions
The Complex-Number Mortality Model (CNMM) based on orthonormal expansion of membership functions

Author(s): Andrzej Szymański, Agnieszka Rossa
Subject(s): Methodology and research technology, Demography and human biology
Published by: Główny Urząd Statystyczny
Keywords: exponential membership functions; Legendre’s polynomials; mortality modelling; orthonormal system;

Summary/Abstract: The paper deals with a new fuzzy version of the Lee-Carter (LC) mortality model, in which mortality rates as well as parameters of the LC model are treated as triangular fuzzy numbers. As a starting point, the fuzzy Koissi-Shapiro (KS) approach is recalled. Based on this approach, a new fuzzy mortality model – CNMM – is formulated using orthonormal expansions of the inverse exponential membership functions of the model components. The paper includes numerical findings based on a case study with the use of the new mortality model compared to the results obtained with the standard LC model.

  • Issue Year: 22/2021
  • Issue No: 3
  • Page Range: 31-57
  • Page Count: 27
  • Language: English