Improved Insights on Financial Health through Partially Constrained Hidden Markov Model Clustering on Loan Repayment Data
25 Jul 2018-Vol. 49, Iss: 3, pp 98-113
TL;DR: This study proposes a modified Hidden Markov Model (HMM) based clustering, which clusters repayment sequences across selected subsets of the HMM parameters, and demonstrates that the ability to cluster on selective parameters, in conjunction with the structural construct of HMMs, enables the discovery of substantially more meaningful business insights than the baselines.
Abstract: There is a growing interest in understanding, as opposed to predicting, the repayment behavior of customers of financial institutions that provide loans. This study proposes a modified Hidden Markov Model (HMM) based clustering, which clusters repayment sequences across selected subsets of the HMM parameters. We demonstrate that different implementations of this modified adaptation helps us gain an in-depth understanding of various drivers that are hard to directly observe but nevertheless govern repayment. These include drivers such as the ability to repay (financial health of the customer) or the intention to repay independent of the ability (willful defaulting and unintentional delinquency). Algorithmically, we achieve this partially constrained HMM clustering (PC-HMM) by placing constraints on the expectation-maximization (EM) algorithm where a subset of parameters are used to cluster the repayments via the estimation process, while the other parameters are learned globally across all repayments. We compare our approach with three other baselines on a real-world loan repayment data set. We use an exogenous variable to validate and benchmark the clusters. We conclude our study with the observation that the ability to cluster on selective parameters, in conjunction with the structural construct of HMMs, enables the discovery of substantially more meaningful business insights than the baselines.
Citations
More filters
TL;DR: In this paper, a systematic review of the literature on artificial intelligence (AI) in customer-facing financial services, providing an overview of explored contexts and research foci, identifying gaps in the literature and setting a comprehensive agenda for future research.
Abstract: The objective of this study is to provide a systematic review of the literature on artificial intelligence (AI) in customer-facing financial services, providing an overview of explored contexts and research foci, identifying gaps in the literature and setting a comprehensive agenda for future research.,Combining database (i.e. Scopus, Web of Science, EBSCO, ScienceDirect) and manual journal search, the authors identify 90 articles published in Australian Business Deans Council (ABDC) journals for investigation, using the TCCM (Theory, Context, Characteristics and Methodology) framework.,The results indicate a split between data-driven and theory-driven research, with most studies either adopting an experimental research design focused on testing the accuracy and performance of AI algorithms to assist with credit scoring or investigating AI consumer adoption behaviors in a banking context. The authors call for more research building overarching theories or extending existing theoretical perspectives, such as actor networks. More empirical research is required, especially focusing on consumers' financial behaviors as well as the role of regulation, ethics and policy concerned with AI in financial service contexts, such as insurance or pensions.,The review focuses on AI in customer-facing financial services. Future work may want to investigate back-office and operations contexts.,The authors are the first to systematically synthesize the literature on the use of AI in customer-facing financial services, offering a valuable agenda for future research.
15 citations
TL;DR: In this article , a detailed examination of the current financial management system's reliability optimization, as well as identifying and supplying the appropriate financial system management tools for the company's business activities is presented.
Abstract: Financial management, as the central link in the enterprise management chain, has emerged as a critical component in achieving the company’s strategic objectives. Businesses are transitioning from a closed to an open and scalable development model to an intensive development model, which is reflected in financial management and necessitates a growth-oriented transformation. Unreliability abounds in today’s globalized environment, from financial management-oriented accounting to efficiency accounting. In addition to general supply and demand unreliability, the high degree of complexity and increased channel unreliability make companies resist the great challenge of financial management systems’ ability to face uncertain events, corporate supply chains become increasingly vulnerable, and unreliability has become a common and indispensable phenomenon in many economic and management decision problems. Due to their high data requirements and dynamic adaptability, Markov models are used as the primary forecasting tool. They do not require a large amount of historical data, but rather a detailed analysis of the recent past. This thesis examines and resolves the issues with the current financial management system through a detailed examination of the enterprise financial management system’s reliability optimization, as well as identifying and supplying the appropriate financial system management tools for the company’s business activities. New business concepts and their advantages to management are introduced. The experimental results show that adding precision constraints to the floating-point calculation problem transforms it into an integer calculation problem, improving reliability by 0.389, demonstrating that the Markov model improves the efficiency of financial resource allocation. At the decision-making level, it provides a management platform and operational tools. As a result, research on this topic will assist businesses in developing a comprehensive business process that meets their needs and adapts to future development needs at a low cost, as well as assisting businesses in communicating with employees, external collaboration, and collaborative business between customers and partners.
01 Jan 2020
TL;DR: In this paper, a database of 100 supporting enterprises from the industry was created according to the SK NACE classification with the representation of the production and services enterprises and the division into the heavy and light industry.
Abstract: Indebtedness is currently the issue that is often addressed in general, which affects the viability and future growth of businesses. It is influenced by many factors that often result from industry specifics. The aim of this contribution is to analyse key indicators of indebtedness in correlation with selected sector-specific indicators of asset structure through an analysis of 100 selected industrial enterprises. For the purpose of the analysis, a database of 100 supporting enterprises from the industry was created according to the SK NACE classification with the representation of the production and services enterprises and the division into the heavy and light industry. These segmentations will be crucial in looking for factors affecting the financial stability of enterprises. We created the database from publicly available financial statements (balance sheet, profit and loss account) published in the FINSTAT database. Key indicators of debt and equity rating were calculated from the prepared data and analysed through a regression and correlation analysis using JMP Pro software support. The results of the analysis point to the increased indebtedness of V4 enterprises compared to the European average, which highlights the need to address this issue.
References
More filters
01 Feb 1989
TL;DR: In this paper, the authors provide an overview of the basic theory of hidden Markov models (HMMs) as originated by L.E. Baum and T. Petrie (1966) and give practical details on methods of implementation of the theory along with a description of selected applications of HMMs to distinct problems in speech recognition.
Abstract: This tutorial provides an overview of the basic theory of hidden Markov models (HMMs) as originated by L.E. Baum and T. Petrie (1966) and gives practical details on methods of implementation of the theory along with a description of selected applications of the theory to distinct problems in speech recognition. Results from a number of original sources are combined to provide a single source of acquiring the background required to pursue further this area of research. The author first reviews the theory of discrete Markov chains and shows how the concept of hidden states, where the observation is a probabilistic function of the state, can be used effectively. The theory is illustrated with two simple examples, namely coin-tossing, and the classic balls-in-urns system. Three fundamental problems of HMMs are noted and several practical techniques for solving these problems are given. The various types of HMMs that have been studied, including ergodic as well as left-right models, are described. >
21,819 citations
10,702 citations
TL;DR: This introduction to the MIS Quarterly Special Issue on Business Intelligence Research first provides a framework that identifies the evolution, applications, and emerging research areas of BI&A, and introduces and characterized the six articles that comprise this special issue in terms of the proposed BI &A research framework.
Abstract: Business intelligence and analytics (BI&A) has emerged as an important area of study for both practitioners and researchers, reflecting the magnitude and impact of data-related problems to be solved in contemporary business organizations. This introduction to the MIS Quarterly Special Issue on Business Intelligence Research first provides a framework that identifies the evolution, applications, and emerging research areas of BI&A. BI&A 1.0, BI&A 2.0, and BI&A 3.0 are defined and described in terms of their key characteristics and capabilities. Current research in BI&A is analyzed and challenges and opportunities associated with BI&A research and education are identified. We also report a bibliometric study of critical BI&A publications, researchers, and research topics based on more than a decade of related academic and industry publications. Finally, the six articles that comprise this special issue are introduced and characterized in terms of the proposed BI&A research framework.
4,610 citations
TL;DR: This article developed several models for limited dependent variables, which are extensions of the multiple probit analysis model and differ from that model by allowing the determination of the size of the variable when it is not zero to depend on different parameters or variables from those determining the probability of its being zero.
Abstract: THIS PAPER DEVELOPS some models for limited dependent variables.2 The distinguishing feature of these variables is that the range of values which they may assume has a lower bound and that this lowest value occurs in a fair number of observations. This feature should be taken into account in the statistical analysis of observations on such variables. In particular, it renders invalid use of the usual regression model. The second section of this paper develops several models for such variables. Like Tobin's [10] model, they are extensions of the multiple probit analysis model.3 They differ from that model by allowing the determination of the size of the variable when it is not zero to depend on different parameters or variables from those determining the probability of its being zero. Estimation and discrimination in the models are considered in Section 3. The models, like their prototypes, seem particularly intractable to exact analysis and large sample approximations have to be used. The adequacy of inferences based on these procedures is explored in Section 4 through a small sampling experiment. Limited dependent variables arise naturally in the study of consumer purchases, particularly purchases of durable goods. When a durable good is to be purchased, the amount spent may vary in fine gradations, but for many durables it is probably the case that most consumers in a particular period make no purchase at all. In Section 5 we apply the models to the demand for durable goods to provide an application of the techniques.
2,808 citations
"Improved Insights on Financial Heal..." refers methods in this paper
...Mofatt (2005) used the double-hurdle model for limited dependent variables (Cragg, 1971) to study the extent of loan default, rather than the probability of default....
[...]