Forecasting credit portfolio components with a Markov chain model

Time series has a wide range of applications in various fields. Recently, a new math tool, named as visibility graph, is developed to transform the time series into complex networks. One shortcoming of the existing network-based time series prediction methods is time consuming. To address this issue, this paper proposes a new prediction algorithm based on visibility graph and Markov chains. Among the existing network-based time series prediction methods, the main step is to determine the similarity degree between two nodes based on link prediction algorithm. A new similarity measure between two nodes is presented without the iteration process in the classical link prediction algorithm. The experiments show that the proposed method has better accuracy with less time consuming.

show abstract

“…In this section, some basic knowledge will be introduced including visibility graph [71], transition probability [66], and Markov chain [72].…”

Section: Preliminariesmentioning

confidence: 99%

A Fast Algorithm for Network Forecasting Time Series

Liu

Deng

2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…• random changes in the financial condition of firms (within the classification: from AAA -the best group to D -default group); • the behavior of an individual borrower in consumer lending [6]. The probability of default, which in many studies is assumed to be equal to the share of problem (overdue) loans, is usually chosen as a loan portfolio risk assessment criteria, In this case, loans are grouped by the existence and terms of debt, probabilities of their transition from one group to another (migration coefficients) are determined, a change in the portfolio structure is modelled, which makes it possible to predict its risk and profitability.…”

Section: Introductionmentioning

confidence: 99%

Mathematical model of the loan portfolio dynamics in the form of Markov chain considering the process of new customers attraction

Bozhalkina

2017

AIP Conference Proceedings

View full text Add to dashboard Cite

Abstract. Mathematical model of the loan portfolio structure change in the form of Markov chain is explored. This model considers in one scheme both the process of customers attraction, their selection based on the credit score, and loans repayment. The model describes the structure and volume of the loan portfolio dynamics, which allows to make medium-term forecasts of profitability and risk. Within the model corrective actions of bank management in order to increase lending volumes or to reduce the risk are formalized.

show abstract

Construction of Confidence Sets for Markov Chain Model

Zavalishchin

Timofeeva

2016

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

Forecasting credit portfolio components with a Markov chain model

Cited by 8 publications

References 8 publications

A Fast Algorithm for Network Forecasting Time Series

A Fast Algorithm for Network Forecasting Time Series

Mathematical model of the loan portfolio dynamics in the form of Markov chain considering the process of new customers attraction

Construction of Confidence Sets for Markov Chain Model

Contact Info

Product

Resources

About