A new approach to deal with variable selection in neural networks: an application to bankruptcy prediction
TL;DR: In this paper, two new procedures that deal with overfitting problem using neural techniques for variable selection and business failure prediction are proposed, which are based simultaneously on backward search, the HVS measure (Heuristic for Variable Selection), and the AUC criterion (Area Under Curve).
Abstract: The purpose of the paper is to propose two new procedures that deal with overfitting problem using neural techniques for variable selection and business failure prediction. The first procedure, called HVS-AUC, is based simultaneously on (i) the backward search, (ii) the HVS measure (Heuristic for Variable Selection), and (iii) the AUC criterion (Area Under Curve). The second procedure, called forward-AUC, is based on (i) the forward search and (ii) the AUC criterion. Using a sample of bankrupt and non-bankrupt firms in France, the implementation of the procedures using neural networks shows that the profitability, the repayment capacity, the taxation, and the importance of investment have a strong explanatory power in bankruptcy prediction. These procedures also provide more parsimonious and more efficient models compared to Linear Discriminant Analysis.
Citations
More filters
2 citations
2 citations
TL;DR: In this article , the authors evaluate machine learning models to predict financial distress in Italian municipalities, using the knowledge of accounting judiciary experts to aid in the early detection of financial distress, leading to better outcomes for the communities.
Abstract: Financial distress of municipalities, although comparable to bankruptcy of private companies, has a far more serious impact on the well-being of communities. For this reason, it is essential to detect deficits as soon as possible. Predicting financial distress in municipalities can be a complex task, as it involves understanding a wide range of factors that can affect a municipality's financial health. In this paper, we evaluate machine learning models to predict financial distress in Italian municipalities. Accounting judiciary experts have specialized knowledge and experience in evaluating the financial performance, and they use a range of indicators to make their assessments. By incorporating these indicators in the feature extraction process, we can ensure that the model is taking into account a wide range of information that is relevant to the financial health of municipalities. The results of this study indicate that using machine learning models in combination with the knowledge of accounting judiciary experts can aid in the early detection of financial distress, leading to better outcomes for the communities.
TL;DR: In this article , the authors proposed and approbate their own approach to determining the level of competitiveness of gas distribution network operators in the western region of Ukraine, taking into account modern trends in the functioning of enterprises.
Abstract: The functioning of Ukrainian national gas sector is directly dependent on the processes of fuel and energy resources consumption and trends in domestic and foreign markets. Nowadays, the majority of approaches and methods are formed with the obligatory use of expert assessment methods, which, in its turn, predetermines relatively subjective judgments and results. In the process of conducting a comprehensive analysis of financial and economic indicators and those reflecting the results of economic activity of gas distribution network operators functioning in the western region of Ukraine, the following approaches have been used in our study with the involvement of: Altman’s two-factor model; Altman’s five-factor model; Lis’s bankruptcy prediction model; Richard Taffler’s model; Beaver’s coefficient; Tereshchenko’s model and Matviychuk’s model; however, the existing models for diagnosing bankruptcy of enterprises are characterized by ambiguity; as for example, if Lis’s model indicates a low bankruptcy level, then other models prove the opposite situation; domestic diagnostic models need to be improved, as they were developed in the early 2000s and disregard current trends in functioning of enterprises. Since the existing models for diagnosing the bankruptcy of enterprises are characterized by ambiguity, the authors proposed and approbate their own approach to determining the level of competitiveness of gas distribution network operators. A feature of the proposed methodology is taking into account modern trends in the functioning of enterprises, taking into account the peculiarities of the activities of gas distribution network operators, and the market stage. A tangible advantage of this approach is the ability to identify the presence or likelihood of critical events at an early stage.
01 Jan 2023
TL;DR: In this paper , the authors take a holistic approach to systematically and qualitatively analyse the feature selection techniques employed by researchers, based on the protocol of the preferred reporting items for systematic reviews and meta-analyses (PRISMA).
Abstract: Systematic literature reviews have long been acknowledged as an important tool for evaluating knowledge theoretically, bridging gaps, and establishing the groundwork for future endeavours. However, in the realm of corporate bankruptcy prediction, systematic literature reviews are few and far between, and none exist for key attributes such as feature selection. This study takes a holistic approach to systematically and qualitatively analyse the feature selection techniques employed by researchers, based on the protocol of the preferred reporting items for systematic reviews and meta-analyses (PRISMA). Furthermore, our goal is to bring together all of the key characteristics, such as feature selection techniques, different machine learning techniques, evaluation criteria, evaluation criteria outcomes, and the research end result, in one place for any interested researcher who wants to do an analysis or view a summary. For the period 2015–2021, the Scopus database and the reference lists of the selected research papers were used to extract a total of 36 articles. The result indicates the split of feature selection approaches under the heading “filter, wrapper, and embedded,” which were employed independently and in combination by various authors. The filter approach is clearly the most preferred amongst researchers due to its simple structure and superior results. Furthermore, we discovered that to acquire the most important variables, multiple feature selection methods were used within the feature selection techniques categories.
References
More filters
TL;DR: In this paper, a set of financial and economic ratios are investigated in a bankruptcy prediction context wherein a multiple discriminant statistical methodology is employed, and the data used in the study are limited to manufacturing corporations, where an initial sample of sixty-six firms is utilized to establish a function which best discriminates between companies in two mutually exclusive groups: bankrupt and nonbankrupt firms.
Abstract: ACADEMICIANS SEEM to be moving toward the elimination of ratio analysis as an analytical technique in assessing the performance of the business enterprise. Theorists downgrade arbitrary rules of thumb, such as company ratio comparisons, widely used by practitioners. Since attacks on the relevance of ratio analysis emanate from many esteemed members of the scholarly world, does this mean that ratio analysis is limited to the world of \"nuts and bolts\"? Or, has the significance of such an approach been unattractively garbed and therefore unfairly handicapped? Can we bridge the gap, rather than sever the link, between traditional ratio \"analysis\" and the more rigorous statistical techniques which have become popular among academicians in recent years? The purpose of this paper is to attempt an assessment of this issue-the quality of ratio analysis as an analytical technique. The prediction of corporate bankruptcy is used as an illustrative case.' Specifically, a set of financial and economic ratios will be investigated in a bankruptcy prediction context wherein a multiple discriminant statistical methodology is employed. The data used in the study are limited to manufacturing corporations. A brief review of the development of traditional ratio analysis as a technique for investigating corporate performance is presented in section I. In section II the shortcomings of this approach are discussed and multiple discriminant analysis is introduced with the emphasis centering on its compatibility with ratio analysis in a bankruptcy prediction context. The discriminant model is developed in section III, where an initial sample of sixty-six firms is utilized to establish a function which best discriminates between companies in two mutually exclusive groups: bankrupt and non-bankrupt firms. Section IV reviews empirical results obtained from the initial sample and several secondary samples, the latter being selected to examine the reliability of the discriminant
10,737 citations
TL;DR: In this paper, the authors present some empirical results of a study predicting corporate failure as evidenced by the event of bankruptcy, and the methodology is one of maximum likelihood estimation of the so-called conditional logit model, in which the data set used in this study is from the seventies (1970-76).
Abstract: This paper presents some empirical results of a study predicting corporate failure as evidenced by the event of bankruptcy. There have been a fair number of previous studies in this field of research; the more notable published contributions are Beaver [1966; 1968a; 1968b], Altman [1968; 1973], Altman and Lorris [1976], Altman and McGough [1974], Altman, Haldeman, and Narayanan [1977], Deakin [1972], Libby [1975], Blum [1974], Edmister [1972], Wilcox [1973], Moyer [1977], and Lev [1971]. Two unpublished papers by White and Turnbull [1975a; 1975b] and a paper by Santomero and Vinso [1977] are of particular interest as they appear to be the first studies which logically and systematically develop probabilistic estimates of failure. The present study is similar to the latter studies, in that the methodology is one of maximum likelihood estimation of the so-called conditional logit model. The data set used in this study is from the seventies (1970-76). I know of only three corporate failure research studies which have examined data from this period. One is a limited study by Altman and McGough [1974] in which only failed firms were drawn from the period 1970-73 and only one type of classification error (misclassification of failed firms) was analyzed. Moyer [1977] considered the period 1965-75, but the sample of bankrupt firms was unusually small (twenty-seven firms). The
5,244 citations
TL;DR: In this article, the authors focus on the use of ratios as predictors of failure, defined as the inability of a firm to pay its financial obligations as they mature, and demonstrate that a firm is said to have failed when any of the following events have occurred.
Abstract: At the turn of the century, ratio analysis was in its embryonic state. It began with the development of a single ratio, the current ratio,' for a single purpose-the evaluation of credit-worthiness. Today ratio analysis involves the use of several ratios by a variety of users-including credit lenders, credit-rating agencies, investors, and management.2 In spite of the ubiquity of ratios, little effort has been directed toward the formal empirical verification of their usefulness. The usefulness of ratios can only be tested with regard to some particular purpose. The purpose chosen here was the prediction of failure, since ratios are currently in widespread use as predictors of failure. This is not the only possible use of ratios but is a starting point from which to build an empirical verification of ratio analysis. "Failure" is defined as the inability of a firm to pay its financial obligations as they mature. Operationally, a firm is said to have failed when any of the following events have occurred: bankruptcy, bond default, an overdrawn bank account, or nonpayment of a preferred stock dividend.3 A "financial ratio" is a quotient of two numbers, where both num-
4,210 citations
TL;DR: In this paper, the authors explored the development of a bankruptcy classification model which incorporates comprehensive inputs with respect to discriminant analysis and utilizes a sample of bankrupt firms essentially covering the period 1969-1975.
Abstract: The paper explores the development of a bankruptcy classification model which incorporates comprehensive inputs with respect to discriminant analysis and utilizes a sample of bankrupt firms essentially covering the period 1969–1975. Financial statement data and market related measures are transformed along guidelines suggested by traditional security analysis to promote comparability of companies and to reflect the most recent reporting standards so as to make the model relevant to future analysis. The results of the study are compared with alternative bankruptcy classification strategies via the explicit introduction of prior probabilities of group membership, observed accuracies, and estimates of costs of errors in misclassification. The latter is based on cost estimates derived from commercial bank lending errors. The results of the study indicate potential significant application to credit worthiness assessment, portfolio management, and to external and internal performance analysis.
1,779 citations
01 Sep 2004
TL;DR: A relatively new machine learning technique, support vector machines (SVM), is introduced to the problem in attempt to provide a model with better explanatory power and relative importance of the input financial variables from the neural network models.
Abstract: Corporate credit rating analysis has attracted lots of research interests in the literature. Recent studies have shown that Artificial Intelligence (AI) methods achieved better performance than traditional statistical methods. This article introduces a relatively new machine learning technique, support vector machines (SVM), to the problem in attempt to provide a model with better explanatory power. We used backpropagation neural network (BNN) as a benchmark and obtained prediction accuracy around 80% for both BNN and SVM methods for the United States and Taiwan markets. However, only slight improvement of SVM was observed. Another direction of the research is to improve the interpretability of the AI-based models. We applied recent research results in neural network model interpretation and obtained relative importance of the input financial variables from the neural network models. Based on these results, we conducted a market comparative analysis on the differences of determining factors in the United States and Taiwan markets.
962 citations