Markov chain sensitivity analysis of expected paid/unpaid overdue receivables – SME case study Cover Image

Markov chain sensitivity analysis of expected paid/unpaid overdue receivables – SME case study
Markov chain sensitivity analysis of expected paid/unpaid overdue receivables – SME case study

Author(s): Ladislav Lukáš
Subject(s): Business Economy / Management, Methodology and research technology
Published by: Masarykova univerzita nakladatelství
Keywords: accounts receivable analysis; fundamental matrix; absorption Markov chains; sensitivity analysis;
Summary/Abstract: The paper uses existing Markov chain theory to estimate expected paid/unpaid overdue receivables, and is focused mainly upon sensitivity analysis of calculated estimations. Since such calculations depend upon fundamental matrix of absorption Markov chain chosen, the particularly important role plays data and algorithm for its composition. As a case study, we selected a SME ranked company which provided us its accounting records with payment pattern details of related receivables. First, the available data are sorted to extract overdue receivables, which serve to estimate transition probability matrices of absorption Markov chains having several transient states and two absorption ones representing paid and unpaid overdue receivables. Based either on number of overdue receivables or their financial volumes we build different transition probability matrices. The sensitivity analysis of expected paid/unpaid overdue receivables concerns influence of different overdue threshold and tolerance accepted, conditional probabilities between transient and absorption states, as well as distribution of financial volumes in particular transient states registered. The results are discussed in detail showing their practical importance in financial management and providing deeper insight into overdue payment processes thus contributing to risk management, too. All computations and graphical issues are performed by sw Mathematica.

  • Page Range: 11-18
  • Page Count: 8
  • Publication Year: 2017
  • Language: English