A neural network approach to the prediction of going concern status

The assessment of a firm's going concern status is not an easy task. To assist auditors, going concern prediction models based on statistical methods such as multiple discriminant analysis and logit/probit analysis have been explored with some success. This study attempts to look at a different and more recent approach—neural networks. In particular, a neural network model of the feedforward, backpropagation type was constructed to predict a firm's going concern status from six financial ratios, using a data set containing 165 non-going concerns and 165 matched going concerns. On an evenly distributed hold-out sample, the trained network model correctly predicted all 30 test cases. The results suggest that neural networks can be a promising avenue of research and application in the going concern area.     

Predicting Bond Ratings Using Neural Networks: A Comparison with Logistic Regression

Bond rating agencies examine the financial outlook of a company and the characteristics of a bond issue and assign a rating that indicates an independent assessment of the degree of default risk associated with the firm’s bonds. Predicting this bond rating has been of interest to potential investors as well as to the firm. Prior research in this area has primarily relied upon traditional statistical methods to develop models with reasonably good prediction accuracy. This article utilizes a neural network approach to modeling the bond rating process in an attempt to increase the overall prediction accuracy of the models. A comparison is made to a more traditional logistic regression approach to classification prediction. The results indicate that the neural networks-based model performs significantly better than the logistic regression model for classifying a holdout sample of newly issued bonds in the 1990–92 period. A potential drawback to a neural network approach is a tendency to overfit the data which could negatively affect the model’s generalizability. This study carefully controls for overfitting and obtains significant improvement in bond rating prediction compared to the logistic regression approach. © 1997 by John Wiley & Sons, Ltd.     

Bankruptcy prediction using fuzzy convolutional neural networks

We propose a combined method for bankruptcy prediction based on fuzzy set qualitative comparative analysis (fsQCA) and convolutional neural networks (CNN). Currently, CNNs are being applied to various fields, and in some areas are providing higher performance than traditional models. In our proposed method, a CNN uses calibrated variables from fuzzy sets to improve performance accuracy. In addition, there are no published studies on the effect of feature selection at the input level of convolutional neural networks. Therefore, this study compares four well-known feature selection methods used in financial distress prediction, (t-test, stepdisc discriminant analysis, stepwise logistic regression and partial least square discriminant analysis) to investigate their effect on classification performance. The results show that fuzzy convolutional neural networks (FCNN) lead to better performance than when using traditional methods.     

Predicting direction shifts on Canadian–US exchange rates with artificial neural networks

The paper presents a variety of neural network models applied to Canadian–US exchange rate data. Networks such as backpropagation, modular, radial basis functions, linear vector quantization, fuzzy ARTMAP, and genetic reinforcement learning are examined. The purpose is to compare the performance of these networks for predicting direction (sign change) shifts in daily returns. For this classification problem, the neural nets proved superior to the naïve model, and most of the neural nets were slightly superior to the logistic model. Using multiple previous days' returns as inputs to train and test the backpropagation and logistic models resulted in no increased classification accuracy. The models were not able to detect a systematic affect of previous days' returns up to fifteen days prior to the prediction day that would increase model performance. Copyright © 2001 John Wiley & Sons, Ltd.     

PREDICTING INDUSTRIAL BOND RATINGS WITH A PROBIT MODEL AND FUNDS FLOW COMPONENTS

Since 1978, there has been a significant change in new bond offerings with a substantive increase in the number of nonconvertible high risk bonds. This study uses an n-chotomous multivariate probit model with cash-based funds flow components and financial ratios to predict industrial bond ratings. The n-chotomous probit model provides superior information for evaluating the bond classification process. The model determines the probabilities of a bond being rated in one of three risk classes. The distribution of the probabilities for each predicted bond rating provides a wealth of new information for evaluating the accuracy of the actual rating. New and reclassified bond ratings by Moody's in 1983 provide the information base for the model that is used to predict 1984 ratings. Initially the classification and predictive results were slightly lower than previous studies. A careful analysis of the probability distributions showed that results were close to being correct in over 90 percent of the cases. The analysis found five cash flow components to be significant in predicting the bond ratings of reclassified issues. The significant components were inventories, other current liabilities, dividends, long-term financing, and fixed coverage charges. The likelihood tests indicated that both ratios and funds flow components contributed information that significantly improved the ability of the n-chotomous multivariate probit model to classify new and revised bond ratings. The study provides valuable insight and nuances concerning the bond-rating process.     

Bankruptcy prediction of financially stressed firms: an examination of the predictive accuracy of artificial neural networks

This is an extension of prior studies that have used artificial neural networks to predict bankruptcy. The incremental contribution of this study is threefold. First, we use only financially stressed firms in our control sample. This enables the models to more closely approximate the actual decision processes of auditors and other interested parties. Second, we develop a more parsimonious model using qualitative ‘bad news’ variables that prior research indicates measure financial distress. Past research has focused on the ‘usefulness’ of accounting numbers and therefore often ignored non‐accounting variables that may contribute to the classification accuracy of the distress prediction models. In addition, rather than use multiple financial ratios, we include a single variable of financial distress using the Zmijewski distress score that incorporates ratios measuring profitability, liquidity, and solvency. Finally, we develop and test a genetic algorithm neural network model. We examine its predictive ability to that of a backpropagation neural network and a model using multiple discriminant analysis. The results indicate that the misclassification cost of the genetic algorithm‐based neural network was the lowest among the models. Copyright © 2001 John Wiley & Sons, Ltd.     

A multiple discriminant analysis of BHC commercial paper ratings

The finance literature contains many examples of attempts to classify bond issuers into agency rating categories and to identify the key variables that contribute to bond rating differences. This study extends the classification literature to include bank holding company (BHC) commercial paper issuers. A multiple discriminant model is developed that effectively classifies paper issuers into their respective Moody's rating groups. An additional discriminant model is derived that classifies these issuers into more detailed ‘market’ rating groups.     

Predicting Corporate Failure Using a Neural Network Approach

This paper investigates the performance of Artificial Neural Networks for the classification and subsequent prediction of business entities into failed and non-failed classes. Two techniques, back-propagation and Optimal Estimation Theory (OET), are used to train the neural networks to predict bankruptcy filings. The data are drawn from Compustat data tapes representing a cross-section of industries. The results obtained with the neural networks are compared with other well-known bankruptcy prediction techniques such as discriminant analysis, probit and logit, as well as against benchmarks provided by directly applying the bankruptcy prediction models developed by Altman (1968) and Ohlson (1980) to our data set. We control the degree of ‘disproportionate sampling’ by creating ‘training’ and ‘testing’ populations with proportions of bankrupt firms ranging from 1% to 50%. For each population, we apply each technique 50 times to determine stable accuracy rates in terms of Type I, Type II and Total Error. We show that the performance of various classification techniques, in terms of their classification errors, depends on the proportions of bankrupt firms in the training and testing data sets, the variables used in the models, and assumptions about the relative costs of Type I and Type II errors. The neural network solutions do not achieve the ‘magical’ results that literature in this field often promises, although there are notable 'pockets' of superior performance by the neural networks, depending on particular combinations of proportions of bankrupt firms in training and testing data sets and assumptions about the relative costs of Type I and Type II errors. However, since we tested only one architecture for the neural network, it will be necessary to investigate potential improvements in neural network performance through systematic changes in neural network architecture.     

Off-site monitoring systems for predicting bank underperformance: a comparison of neural networks,discriminant analysis,and professional human judgment

This study compares the ability of discriminant analysis, neural networks, and professional human judgment methodologies in predicting commercial bank underperformance. Experience from the banking crisis of the 1980s and early 1990s suggest that improved prediction models are needed for helping prevent bank failures and promoting economic stability. Our research seeks to address this issue by exploring new prediction model techniques and comparing them to existing approaches. When comparing the predictive ability of all three models, the neural network model shows slightly better predictive ability than that of the regulators. Both the neural network model and regulators significantly outperform the benchmark discriminant analysis model's accuracy. These findings suggest that neural networks show promise as an off-site surveillance methodology. Factoring in the relative costs of the different types of misclassifications from each model also indicates that neural network models are better predictors, particularly when weighting Type I errors more heavily. Further research with neural networks in this field should yield workable models that greatly enhance the ability of regulators and bankers to identify and address weaknesses in banks before they approach failure. Copyright © 2001 John Wiley & Sons, Ltd.     

Classification techniques for the identification of falsified financial statements: a comparative analysis

In this study, I develop 10 alternative classification models using logit analysis, discriminant analysis, support vector machines, artificial neural networks, probabilistic neural networks, nearest neighbours, UTADIS and MHDIS for the detection of falsified financial statements. The models are developed using financial and nonfinancial data. The sample includes 398 financial statements, half of which were assigned a qualified audit opinion. I compare these alternatives methods using out‐of‐time and out‐of‐sample tests. The results are used to derive conclusions on the performance of the methods and to investigate the potential of developing models that will assist auditors in identifying fraudulent financial statements. Copyright © 2009 John Wiley & Sons, Ltd.     

THE EFFECT OF BOND RATING AGENCIES ON BOND RATING MODELS

This paper addresses the problem of bond rating discrepancies and their effect on bond rating prediction models. Both Moody's and Standard & Poor's now use modified ratings. Results of this study indicate that the two agencies disagree 58 percent of the time and that Moody's rates bonds significantly lower than S & P. In addition, the classification rates of the multiple discriminant analysis models decrease approximately 24 percentage points when the modified ratings are used.     

Dynamic aspects in neural classification

When using neural networks for classification in an environment that is changing as time proceeds, methods for updating the parameters of the neural network should be considered in order to retain classification accuracy. Otherwise the neural network may lose performance due to structural changes. A classifier could be completely relearned from scratch at regular intervals. However, our experience from past credit scoring applications shows that users commonly prefer systems that change in as few cases as possible. Furthermore, this approach may be wasteful regarding the required computing time and that previously learned information will always be discarded. Therefore, we favor a methodology that attempts to detect changes and adapts a classifier only if inevitable. In this article, some methods for detecting and treating structural changes are applied to a credit scoring application. The results show that these methods may successfully be applied in a dynamic setting. Copyright © 1999 John Wiley & Sons, Ltd.     

Comparative performance of backpropagation networks designed by genetic algorithms and heuristics

The design of neural network models involves numerous complexities, including the determination of input vectors, choosing the number of hidden layers and their computational units, and specifying activation functions for the latter. The combinatoric possibilities are daunting, yet experience has yielded informal guidelines that can be useful. Alternatively, current research on genetic algorithms (GA) suggests that they might be of practical use as a formal method of determining ‘good’ architectures for neural networks. In this paper, we use a genetic algorithm to find effective architectures for backpropagation neural networks (BP). We compare the performance of heuristically designed BP networks with that of GA-designed BP networks. Our test domains are sets of problems having compensatory, conjunctive, and mixed-decision structures. The results of our experiment suggest that heuristic methods produce architectures that are simpler and yet perform comparatively well. © 1998 John Wiley & Sons, Ltd.     

Introducing Recursive Partitioning for Financial Classification: The Case of Financial Distress

The purpose of this study is to present a new classification procedure, Recursive Partitioning Algorithm (RPA), for financial analysis and to compare it with discriminant analysis within the context of firm financial distress. RPA is a computerized, nonparametric technique based on pattern recognition which has attributes of both the classical univariate classification approach and multivariate procedures. RPA is found to outperform discriminant analysis in most original sample and holdout comparisons. We also observe that additional information can be derived by assessing both RPA and discriminant analysis results.     

Performance of Neural Networks in Managerial Forecasting

This paper investigates the effectiveness of a multi-layered neural network as a tool for forecasting in a managerial time-series setting. To handle noisy data of limited length we adopted two different neural network approaches. First, the neural network is used as a pattern classifier to automate the ARMA model-identification process. We tested the performance of multi-layered neural networks with two statistical feature extractors: ACF/PACF and ESACF. We found that ESACF provides better performance, although the noise in ESACF patterns still caused the classification performance to deteriorate. Therefore we adopted the noise-filtering network as a preprocessor to the pattern-classification network, and were able to achieve an average of about 89% classification accuracy. Second, the neural network is used as a tool for function approximation and prediction. To alleviate the overfitting problem we adopted the structure of minimal networks and recurrent networks. The experiment with three real-world time series showed that the prediction by Elman's recurrent network outperformed those by the ARMA model and other structures of multi-layered neural networks, especially when the time series contained significant noise.     

A comparison of the relative costs of financial distress models: artificial neural networks,logit and multivariate discriminant analysis

This study uses two artificial neural networks (ANNs), categorical learning/instar ANNs and probabilistic (PNN) ANNs, suitable for classification and prediction type issues, and compares them to traditional multivariate discriminant analysis (MDA) and logit to examine financial distress one to three years prior to failure. The results indicate that traditional MDA and logit perform best with the lowest overall error rates. However, when the relative error costs are considered, the ANNs perform better than traditional logit or MDA. Also, as the time period moves farther away from the eventual failure date, ANNs perform more accurately and with lower relative error costs than logit or MDA. This supports the conclusion that for auditors and other evaluators interested in early warning techniques, categorical learning network and probabilistic ANNs would be useful. © 1997 John Wiley & Sons, Ltd.     

MULTIVARIATE ANALYSIS OF CORPORATE BOND RATINGS AND INDUSTRY CLASSIFICATIONS

Previous statistical models designed to predict bond ratings use a variety of financial data and ratios; however, the possible effect of the industry in which the firms operate is generally disregarded. Although the exact bond rating process used by the rating agencies is not known, industry analysis is an important component of the process. This study evaluates the use of industry-specific multiple discriminant functions for predicting bond ratings. The results support using such a technique.     

Accurate Short-Term Yield Curve Forecasting using Functional Gradient Descent

We propose a multivariate nonparametric technique for generatingreliable short-term historical yield curve scenarios and confidenceintervals. The approach is based on a Functional Gradient Descent(FGD) estimation of the conditional mean vector and covariancematrix of a multivariate interest rate series. It is computationallyfeasible in large dimensions and it can account for nonlinearitiesin the dependence of interest rates at all available maturities.Based on FGD we apply filtered historical simulation to computereliable out-of-sample yield curve scenarios and confidenceintervals. We back-test our methodology on daily USD bond datafor forecasting horizons from 1 to 10 days. Based on severalstatistical performance measures we find significant evidenceof a higher predictive power of our method when compared toscenarios generating techniques based on (i) factor analysis,(ii) a multivariate CCC-GARCH model, or (iii) an exponentialsmoothing covariances estimator as in the RiskMetrics^TM approach.     

Features selection,data mining and finacial risk classification: a comparative study

The aim of this paper is to compare several predictive models that combine features selection techniques with data mining classifiers in the context of credit risk assessment in terms of accuracy, sensitivity and specificity statistics. The t‐statistic, Battacharrayia statistic, the area between the receiver operating characteristic, Wilcoxon statistic, relative entropy, and genetic algorithms were used for the features selection task. The selected features are used to train the support vector machine (SVM) classifier, backpropagation neural network, radial basis function neural network, linear discriminant analysis and naive Bayes classifier. Results from three datasets using a 10‐fold cross‐validation technique showed that the SVM provides the best accuracy under all features selections techniques adopted in the study for all three datasets. Therefore, the SVM is an attractive classifier to be used in real applications for bankruptcy prediction in corporate finance and financial risk management in financial institutions. In addition, we found that our best results are superior to earlier studies on the same datasets.     

Measuring Corporate Bond Mortality and Performance

This study develops an alternative way to measure default risk and suggests an appropriate method to assess the performance of fixed-income investors over the entire spectrum of credit-quality classes. The approach seeks to measure the expected mortality of bonds and the consequent loss rates in a manner similar to the way actuaries assess mortality of human beings. The results show that all bond ratings outperform riskless Treasuries over a ten-year horizon and that, despite relatively high mortality rates, B-rated and CCC-rated securities outperform all other rating categories for the first four years after issuance, with BB-rated securities outperforming all others thereafter.