Modelling corporate failure: A multinomial nested logit analysis for unordered outcomes

This study evaluates the theoretical and empirical significance of the multinomial nested logit (NL) model as an advanced closed-form model for the explanation and prediction of firm financial distress. Using a four-state failure model based on Australian company samples, we estimate an NL model and test its predictive performance on a holdout sample. Comparison of model fits and out-of-sample forecasts indicate that the unordered NL model statistically outperforms a standard logit model by substantial margins. NL may even be used as an effective practical alternative to more advanced open-form models such as mixed logit in the modelling of firm financial distress.     

系统性风险与企业财务危机预警——基于前沿机器学习的新视角

本文采用Logit回归模型以及随机森林模型、梯度提升模型等前沿机器学习方法,深入考察系统性风险指标对我国企业财务危机的预测能力。结果表明,系统性风险对中下游企业的财务危机具有显著的预测能力,而基于因子分析构建的系统性风险指标,结合随机森林模型可取得更好的预测效果。本文进一步区分财务危机的不同成因并发现,基于随机森林模型和Logit回归模型的预测框架能够对我国大多数财务危机事件进行有效预警。在此基础上,本文对我国上市企业监管提出相关建议,从而为完善金融风险处置机制提供一定参考。     

Forecasting Corporate Bankruptcy: Optimizing the Performance of the Mixed Logit Model

In recent studies, Jones and Hensher (2004 , 2005) provide an illustration of the usefulness of advanced probability modelling in the prediction of corporate bankruptcies, insolvencies and takeovers. Mixed logit (or random parameter logit) is the most general of these models and appears to have the greatest promise in terms of underlying behavioural realism, desirable econometric properties and overall predictive performance. It suggests a number of empirical considerations relevant to harnessing the maximum potential from this new model (as well as avoiding some of the more obvious pitfalls associated with its use). Using a three-state failure model, the unconditional triangular distribution for random parameters offers the best population-level predictive performance on a hold-out sample. Further, the optimal performance for a mixed logit model arises when a weighted exogenous sample maximum likelihood (WESML) technique is applied in model estimation. Finally, we suggest an approach for testing the stability of mixed logit models by re-estimating a selected model using varying numbers of Halton intelligent draws. Our results have broad application to users seeking to apply more accurate and reliable forecasting methodologies to explain and predict sources of firm financial distress better.     

An Empirical Investigation of Firm Longevity: A Model of the Ex Ante Predictors of Financial Distress

Empirical models of a potential failure process that incorporate distress states between the extremes of corporate health and bankruptcy are uncommon. We depict financial distress as a series of financial events that reflect varied stages of corporate adversity. Our intent is to provide information regarding the influence of certain risk dimensions and firm-specific attributes on distressed firm survival over time. Within a theorized distress framework, we utilize the techniques of survival analysis to longitudinally track firms, grouped a priori according to an initial decline in operating cash flows. We find that the event of default has a significant positive association with business failure. Further, we document that the significant accounting covariates tend to change conditional on a firm having progressed through the diverse stages of distress. These findings accentuate the heterogeneous nature of financial distress and potential business failure.     

Detecting stock market manipulation via machine learning: Evidence from China Securities Regulatory Commission punishment cases

In this paper, we apply machine-learning techniques to construct detecting models of stock market manipulation. By combining manually collected China Securities Regulatory Commission punishment cases from 2014 to 2016 with financial information of listed companies, we construct a training set and a test set to compare the detecting ability of support vector machine (SVM) and logistic model. Considering imbalanced data, we further incorporate Borderline Synthetic Minority Oversampling Technique (Borderline SMOTE) to oversample minority class and then find that Borderline SMOTE–SVM performs better than SVM and benchmark model in detecting manipulation. To enhance detecting performance of the models, we innovatively introduce market sentiment indicators which are extracted from analyst rating reports, financial news, and Guba comments into our indicators set. The results indicate that the new indicators generate significant marginal increment to the model accuracy.     

Research on optimization of an enterprise financial risk early warning method based on the DS-RF model

The financial risk early warning process of enterprises faces problems such as uncertainty and complexity. In the big data environment, scholars and enterprises that continue to use traditional evaluation methods will face large challenges. It is essential for an enterprise's sustainable operation to combine artificial intelligence algorithms, dynamically monitor its financial risks, and carry out financial risk early warning processes accurately and effectively. This study proposes an early warning method for corporate financial risks based on the evidence theory-random forest (DS-RF) model. The classic algorithm of machine learning—random forest was introduced into the framework of evidence theory to construct a random forest model with four dimensions: profitability, asset quality, debt risk, and operating growth. While predicting the risk, the credibility of the evidence was determined, and then the D-S synthesis rule was used for information fusion. An example was analyzed, taking JS Reclamation Group as the study subject. The comparison with the early warning results of the random forest algorithm and the traditional model shows that the DS-RF model proposed in this paper has a higher early warning accuracy and the results are presented more comprehensively and systematically, which effectively improves the efficiency of enterprise financial risk early warning and helps managers to make relevant decisions efficiently and scientifically.     

Factors Affecting the Probability of Bankruptcy: A Managerial Decision Based Approach

The majority of classification models developed have used a pool of financial ratios combined with statistical variable selection techniques to maximize the accuracy of the classifier constructed. Rather than follow this approach, this article seeks to provide an explicit economic basis for the selection of variables for inclusion in bankruptcy models. This search to develop an economic theory of bankruptcy augments the existing bankruptcy prediction literature. Variables which occur in bankruptcy probability expressions derived from the solution of a stochastic optimizing model of firm behaviour are 'proxied' by variables constructed from financial statement data. The random nature of the lifetime of a single firm provides the rationale for the use of duration or hazard-based statistical methods in the validation of the derived bankruptcy probability expressions. Results of the validation exercise confirm that the majority of variables included in the empirical hazard formulation behave in a way that is consistent with the model of the firm. The results highlight the need for developments in the measurement of earnings dispersion.     

Predicting credit card fraud with Sarbanes‐Oxley assessments and Fama‐French risk factors

This research developed and tested machine learning models to predict significant credit card fraud in corporate systems using Sarbanes‐Oxley (SOX) reports, news reports of breaches and Fama‐French risk factors (FF). Exploratory analysis found that SOX information predicted several types of security breaches, with the strongest performance in predicting credit card fraud. A systematic tuning of hyperparamters for a suite of machine learning models, starting with a random forest, an extremely‐randomized forest, a random grid of gradient boosting machines (GBMs), a random grid of deep neural nets, a fixed grid of general linear models where assembled into two trained stacked ensemble models optimized for F1 performance; an ensemble that contained all the models, and an ensemble containing just the best performing model from each algorithm class. Tuned GBMs performed best under all conditions. Without FF, models yielded an AUC of 99.3% and closeness of the training and validation matrices confirm that the model is robust. The most important predictors were firm specific, as would be expected, since control weaknesses vary at the firm level. Audit firm fees were the most important non‐firm‐specific predictors. Adding FF to the model rendered perfect prediction (100%) in the trained confusion matrix and AUC of 99.8%. The most important predictors of credit card fraud were the FF coefficient for the High book‐to‐market ratio Minus Low factor. The second most influential variable was the year of reporting, and third most important was the Fama‐French 3‐factor model R² – together these described most of the variance in credit card fraud occurrence. In all cases the four major SOX specific opinions rendered by auditors and the signed SOX report had little predictive influence.     

Hierarchical determinants of capital structure

We analyze the influence of time-, firm-, industry- and country-level determinants of capital structure. First, we apply hierarchical linear modeling in order to assess the relative importance of those levels. We find that time and firm levels explain 78% of firm leverage. Second, we include random intercepts and random coefficients in order to analyze the direct and indirect influences of firm/industry/country characteristics on firm leverage. We document several important indirect influences of variables at industry and country-levels on firm determinants of leverage, as well as several structural differences in the financial behavior between firms of developed and emerging countries.     

Consumer credit-risk models via machine-learning algorithms

We apply machine-learning techniques to construct nonlinear nonparametric forecasting models of consumer credit risk. By combining customer transactions and credit bureau data from January 2005 to April 2009 for a sample of a major commercial bank’s customers, we are able to construct out-of-sample forecasts that significantly improve the classification rates of credit-card-holder delinquencies and defaults, with linear regression R²’s of forecasted/realized delinquencies of 85%. Using conservative assumptions for the costs and benefits of cutting credit lines based on machine-learning forecasts, we estimate the cost savings to range from 6% to 25% of total losses. Moreover, the time-series patterns of estimated delinquency rates from this model over the course of the recent financial crisis suggest that aggregated consumer credit-risk analytics may have important applications in forecasting systemic risk.     

A MULTIPERIOD MEAN-VARIANCE MODEL OF OPTIMAL CAPITAL STRUCTURE

The continuing interest in the capital structure issue among financial researchers is evidenced by the stream of capital structure models that have appeared in the literature. Much of this research has used a risk-neutral and/or a single-period framework. In this paper, we develop a capital structure model for multiperiod firms and allow for the firm's cash flows to grow over time, for the firm to issue new debt, and for two types of bankruptcy costs to occur. The types of bankruptcy costs that occur are determined by the firm's uncertain operating cash flows and negotiations between the firm and creditors. Risk is priced via the Sharpe-Lintner capital asset pricing model. Multiperiod risk-priced models, we argue, realistically represent actual firms and are thus an important step toward the development of more testable and usable models of capital structure. We execute a demonstration example in which the value of the levered firm achieves a maximum and discuss the steps the firm would take to maximize shareholder wealth within this example. The example illustrates that the value of the firm passes through an interior optimum as the promised debt payment is increased. A simulation of the effect of changes in firm-specific parameters shows that the model exhibits expected and appealing relationships between these parameters and the value of the levered firm.     

DEA as a business failure prediction tool Application to the case of galician SMEs

In the research group we are working to provide further empirical evidence on the business failure forecast. Complex fitting modelling; the study of variables such as the audit impact on business failure; the treatment of traditional variables and ratios have led us to determine a starting point based on a reference mathematical model. In this regard, we have restricted the field of study to non-financial galician SMEs in order to develop a model¹ to diagnose and forecast business failure. We have developed models based on relevant financial variables from the perspective of the financial logic, voltage and financial failure, applying three methods of analysis: discriminant, logit and multivariate linear. Finally, we have closed the first cycle using mathematical programming –DEA or Data Envelopment Analysis– to support the failure forecast. The simultaneous use of models was intended to compare their respective conclusions and to look for inter-relations. We can say that the resulting models are satisfactory on the basis of their capacity for prediction. Nevertheless, DEA contains significant points of criticism regarding its applicability to business failure.     

IPO Prospectus Information and Subsequent Performance

Initial public offerings underperform in the long run; however, there is very little evidence on their cross‐sectional variation. Using a random sample of IPOs from 1987 through 1991 and gathering their prospectus data, we show that financial and operating characteristics as well as offering characteristics have a limited relation with the one‐year stock returns. We also find that firms that subsequently reissue equity or merge outperform their matched‐firm benchmarks over three years. Underperformance is most severe for the smaller and younger firms. We find that prospectus information is more useful to predict survival/failure compared to subsequent equity offerings or acquisitions.     

Can alert models for fraud protect the elderly clients of a financial institution?

ABSTRACT

Using account-level transaction data at a major financial institution, we predict the incidence of suspicious activity that can be related to the external financial fraud of its elderly clients. The data consists of over 5 million accounts of clients aged 70 years and older, and over 250 million transactions extending from January 2015 to August 2016. Our main focus is to improve the detection of alerts within a proprietorial transaction monitoring system. Using logistic regression, random forest and support vector machine learning techniques, together with corrections for imbalanced alert samples, we provide a new alert model for the protection of elderly clients at a financial institution, with out-of-sample predictive accuracy. Our findings show the relative influence of client traits and account activity in our select external fraud alert models.     

Do narrative-related disclosures predict corporate failure? Evidence from UK non-financial publicly quoted firms

By creating a comprehensive corporate failure-related lexicon, this paper explores the incremental explanatory power of narrative-related disclosures in predicting corporate failure. We find that corporate failure-related narrative disclosures significantly predict firms' failure up to two years ahead of actual failure. Additionally, we find that a financially distressed firm would become more vulnerable when financial constraints befall, which in turn would precipitate corporate failure. Various robustness tests assure the credibility of the explanatory ability of corporate failure-related narrative disclosures to predict corporate failure. Collectively, our results show the feasibility of these narrative-related disclosures in improving the explanatory power of models that predict corporate failure.     

VaRSOM: A Tool to Monitor Markets' Stability Based on Value at Risk and Self-Organizing Maps

We introduce a variant of self-organizing maps (SOMs) termed VaRSOM that evaluates the similarity among inputs and nodes of the map employing value at risk (VaR). In this way we embed risk measurement within a machine-learning architecture, thus becoming particularly well-suited to analysing financial data. We tested the visualization capabilities and the explicative power of VaRSOM on data from the German Stock Exchange; we then evaluated the results in a comparative perspective, opposing the VaRSOM outcomes to those of SOM trained with more conventional similarity measures. The results lead to the conclusion that VaRSOM is a tool particularly well suited to visualize and exploit critical patterns in financial markets. This, in turn, opens perspectives for a general machine-learning framework sensitive to financial distress and contagion effects. Copyright © 2015 John Wiley & Sons, Ltd.     

Proportionate Growth and the Theoretical Foundations of Financial Ratios

The article proposes a theoretical framework for understanding financial ratios, showing that the multiplicative character of the financial variables from which financial ratios are constructed is a necessary condition of valid ratio usage, not just an assumption supported by evidence. Also, by assuming that firm size is a measurable statistical effect, the article offers an informed reappraisal of the limitations of financial ratios, particularly the well–known limitation of proportionality. The article is divided into two parts, one where ratio components are viewed as deterministic vari– ables and the other where they are random. Such an approach allows the characteristics of ratios to be more easily understood before generalizing the relationship between ratio components to encompass randomness. In the second part, when variability introduced by firm size is treated as a random effect, it is shown that if the accounting variables Y and X used to calculate a financial ratio Y/X are exponential Brownian motion, and if continuous growth rates are equal and proportionate to firm size, this may lead to ratios which are asymmetric but which do not necessarily drift.     

A comparison of alternative bankruptcy prediction models

Early models of bankruptcy prediction employed financial ratios drawn from pre-bankruptcy financial statements and performed well both in-sample and out-of-sample. Since then there has been an ongoing effort in the literature to develop models with even greater predictive performance. A significant innovation in the literature was the introduction into bankruptcy prediction models of capital market data such as excess stock returns and stock return volatility, along with the application of the Black–Scholes–Merton option-pricing model. In this note, we test five key bankruptcy models from the literature using an up-to-date data set and find that they each contain unique information regarding the probability of bankruptcy but that their performance varies over time. We build a new model comprising key variables from each of the five models and add a new variable that proxies for the degree of diversification within the firm. The degree of diversification is shown to be negatively associated with the risk of bankruptcy. This more general model outperforms the existing models in a variety of in-sample and out-of-sample tests.     

Towards an understanding of profitability analysis within the residual income valuation framework

This paper argues that there is a mis-match between formal theoretical accounting valuation models, and practical approaches to profitability analysis and valuation. In particular, none of the linear information models published to date exhibit an obvious role for profitability analysis. For example, in the standard Ohlson model, earnings and book value apparently summarise all the value relevant information available from the firm's financial statements and there is no apparent need for any further investigation of the accounting numbers beyond these specific line items. The purpose of this paper is to attempt to investigate potential analytical links between formal valuation models and practical profitability analysis. Specifically, we attempt to show how key features of practical profitability analysis might be incorporated into formal valuation models. In this respect there are two particular aspects of valuation practice to which the formal models published to date have paid no attention. First, in practice we often see explicit reference made to the demand side (sales), and supply side (costs) of the business. Second, we often see attempts to benchmark the financial ratios of one firm against the corresponding ratios of firms in the same industry. The purpose of this paper is to attempt to explain why such practices make sense in the context of an attempt to model the principal determinants of firm value within a residual income valuation framework.     

CAP Forum on E‐Business: The Management of Financial Disclosure on Corporate Websites: A Conceptual Model*

This paper addresses the impact of Internet financial reporting (IFR) on financial accounting theory by incorporating it into the general Gibbins, Richardson, and Waterhouse (GRW) (1990) disclosure‐management framework. The GRW model assumes that the firm has a relatively stable process of disclosure management. This process varies between two positions: one ritualistic and the other opportunistic. These dimensions can coexist in the same firm but, on average, the policy of a firm will be either more ritualistic or more opportunistic. Our survey of the financial information disclosed in traditional financial reporting (TFR) as compared with the website disclosures of a random sample of Canadian companies documents a significant difference between TFR and IFR, as well as a wide variability among the sample firms in their use of IFR content, format, and technology. We interpret this variability in the incremental difference of IFR over TFR as an indication that a firm's ritualistic or opportunistic behaviour under IFR is not different from its behaviour under TFR. Thus, the adapted GRW (1990) conceptual model appears to have the potential to support future research in the management of financial disclosure on corporate websites.