Belief Networks for Expert System Development in Auditing

This study examines the use of a belief network based expert system for an auditing task—financial distress evaluation for banks. A belief network uses probability measures to store important dependencies across variables of interest in a problem domain, and makes inferences based on observed evidence using probability calculus. This paper discusses how belief network structures can be constructed, and used to assist auditor's in making appropriate recommendations regarding the financial health of a bank under audit. The ability of a belief network to make reliable predictions depends on how well the network structure reflects the underlying dependencies across variables in the problem domain (e.g. financial ratios and the financial health of a bank). The first part of this study illustrates how a computer program developed by the authors can be used to generate and evaluate different feasible belief network structures based on historical data. The program uses an information-theoretic measure to compare the alternative structures. The ability of the program to identify existing dependencies across variables is demonstrated by using it to reconstruct a known network structure from simulated data. Next, the program is used on a database of twelve important bank financial ratios over a three-year period. The predictive ratios identified by the program reflect important areas of a bank's health, such as loan quality, efficiency, profitability and capital adequacy. Finally, a belief revision mechanism is encoded for the belief network structure identified earlier, and is used to illustrate how it can assist auditors in making recommendations about financial health based on a bank's critical financial ratios. The probability estimates provided by the system are validated using data on banks not used in the network design stage, and are found to be reliable.     

An exploratory study of factors affecting credit card brand and category selection in India

This article aims to identify various credit card selection factors in the context of an Asian emerging economy – India. Considering the recent emergence of the three-tier banking system in India, the study proposes to investigate whether there exists an association between credit card selection factors and the type of credit card issuing bank (public, private national or private foreign). The article also seeks to find out whether the income level of the individual determines the type of credit card bank category selection, as prior literature suggests that demographic factors are closely related to credit card selection and usage patterns. The analysis was carried out using a mixed method research design involving thematic analysis of focus group data, factor analysis, PERMAP analysis and multiple correspondence analysis in different phases. Five primary credit card selection factors were identified. A strong correspondence between credit card selection factors, personal income levels and choice of credit card issuing bank was found. Accordingly, credit card selection factors and personal income level were together identified as factors affecting the type of bank selected. The phenomenon of bank category-based stereotyping of brands is emphasized based on the analysis. The implications of this phenomenon with respect to brand identity conception, target segmentation, brand positioning and marketing communication are discussed.     

Contingent means in multi-life models

Multi-life models are useful in actuarial science for studying life contingency. Contingent probabilities are well-understood by most actuaries and are discussed extensively in the existing actuarial literature. However, the mean of a life in a multi-life model involving order of deaths is often found to be rather challenging to interpret by most actuaries who do not understand measure-theoretic probability. Standard textbooks on actuarial science or statistics do not elaborate on the correct interpretation of contingent means, leaving the actuaries at risk of making a blunder. This paper presents the correct interpretation both heuristically and rigorously using a non-measure-theoretic language, so that actuaries will be aware of some common misconceptions and avoid pitfalls in their work. The primary audience of this paper is practicing actuaries, actuarial students and actuarial educators. So we have given several actuarial applications. We hope that applied statisticians also will find this paper useful.     

Assessing the long‐run economic impact of labour law systems: a theoretical reappraisal and analysis of new time series data

Standard economic theory sees labour law as an exogenous interference with market relations and predicts mostly negative impacts on employment and productivity. We argue for a more nuanced theoretical position: labour law is, at least in part, endogenous, with both the production and the application of labour law norms influenced by national and sectoral contexts, and by complementarities between the institutions of the labour market and those of corporate governance and financial markets. Legal origin may also operate as a force shaping the content of the law and its economic impact. Time‐series analysis using a new data set on legal change from the 1970s to the mid‐2000s shows evidence of positive correlations between regulation and growth in employment and productivity, at least for France and Germany. No relationship, either positive or negative, is found for the UK and, although the United States shows a weak negative relationship between regulation and employment growth, this is offset by productivity gains.     

QaaS (quality as a service) model for web services using big data technologies

Quality of service (QoS) determines the service usability and utility and both of which influence the service selection process. The QoS varies from one service provider to other. Each web service has its own methodology for evaluating QoS. The lack of transparent QoS evaluation model makes the service selection challenging. Moreover, most QoS evaluation processes do not consider their historical data which not only helps in getting more accurate QoS but also helps for future prediction, recommendation and knowledge discovery. QoS driven service selection demands a model where QoS can be provided as a service to end users. This paper proposes a layered QaaS (quality as a service) model in the same line as PaaS and software as a service, where users can provide QoS attributes as inputs and the model returns services satisfying the user’s QoS expectation. This paper covers all the key aspects in this context, like selection of data sources, its transformation, evaluation, classification and storage of QoS. The paper uses server log as the source for evaluating QoS values, common methodology for its evaluation and big data technologies for its transformation and analysis. This paper also establishes the fact that Spark outperforms the Pig with respect to evaluation of QoS from logs.     

Attracting private investment: Tax reduction,investment subsidy,or both?

This paper uses a real-option model to examine the net benefit to a government from using tax cut and/or investment subsidy as incentives to induce immediate investment. Although earlier papers generally concluded that investment subsidy dominates tax cut, it is observed that many governments use a combination of subsidy and tax cut. We show that, when the government uses a different discount rate from private firms, and when it has to borrow money to provide an investment subsidy, it is possible to get an internal optimum; that is, it might be optimal for the government to provide an investment subsidy as well as charge a positive tax rate on the profits from the project. Thus, we provide an explanation for the puzzling fact that many governments provide an investment subsidy to a firm while simultaneously taxing its profits.     

Analysis of binary data with the possibility of wrong ascertainment

Most genetic studies recruit high‐risk families, and the discoveries are based on non‐random selected groups. We must consider the consequences of this ascertainment process to apply the results of genetic research to the general population. In addition, in epidemiological studies, binary responses are often misclassified. We proposed a binary logistic regression model that provides a novel and flexible way to correct for misclassification in binary responses, taking into account the ascertainment issues. A hierarchical Bayesian analysis using Markov chain Monte Carlo method has been carried out to investigate the effect of covariates on disease status. The focus of this paper is to study the effect of classification errors and non‐random ascertainment on the estimates of the model parameters. An extensive simulation study indicated that the proposed model results in substantial improvement of the estimates. Two data sets have been revisited to illustrate the methodology.     

On Delays in Project Completion with Cost Reduction: An Experiment

We examine the voluntary provision of a public project via binary contributions when contributions may be made over multiple periods. In many situations, early contributors are likely to pay a higher cost than those who wait. We show that in such circumstances the provision of the project always involves delay. Because this game involves coordination on complex, dynamic strategies in the face of asymmetries in payoffs, we examine behavior in the laboratory.