Stewart Jones

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 Error Component Logit Analysis of Corporate Bankruptcy and Insolvency Risk in Australia

This paper introduces a four-state failure model to depict a wider range of distress scenarios that public companies typically face in the real world. We use a multinomial error component logit model to analyse firm failure, a major advance on the modelling techniques used in previous research. The error component logit model, being an extension of the more familiar mixed logit model, relaxes several questionable statistical assumptions associated with standard models. Using a sample of Australian firms we provide an interpretative illustration of the error component logit model and contrast its behavioural performance with the standard logit model widely used in previous research.

The impact of free cash flow, financial leverage and accounting regulation on earnings management in Australia’s “old” and “new” economies

Outlines the rapid growth of “new economy” companies in Australia and compares their levels of earnings management with “old economy” firms, using data on all Australian listed companies. Reviews the relevant research, explains the methodology and presents the results. Shows that the old economy firms do engage in significant earnings management which is positively associated with leverage and free cash flow levels but, surprisingly, that this is far less evident in the new economic sector. Considers consistency with other research, the underlying reasons for the findings (including regulatory constraints) and opportunities for further research.

Market Behaviour around Bankruptcy Announcements: Evidence from the Australian Stock Exchange

The corporate distress literature to date has largely focused on the predictive power of accounting variables (Altman, 2001). Following previous literature, this study examines the relevance of abnormal stock returns in discriminating between failed and non-failed firms (e.g. Clark and Weinstein, 1983; Shumway, 2001). Our results confirm the findings of previous literature that investors in failed firms typically incur substantial negative stock returns leading up to failure announcements. However, in contrast to prior research we do not find evidence of an announcement effect (i.e. negative stock returns on the event day itself or the day preceding). We also document evidence that the bid-ask spreads of failed firms widen substantially up to 7 months prior to failure, indicating the likelihood of significant information asymmetries across investors in failed firms.

Evaluating the Behavioural Performance of Alternative Logit Models: An Application to Corporate Takeovers Research

Econometric models involving a discrete outcome dependent variable abound in the finance and accounting literatures. However, much of the literature to date utilises a basic or standard logit model. Capitalising on recent developments in the discrete choice literature, we examine three advanced (or non-IID) logit models, namely: nested logit, mixed logit and latent class MNL. Using an illustration from corporate takeovers research, we compare the explanatory and predictive performance of each class of advanced model relative to the standard model. We find that in all cases the more advanced logit model structures, which correct for the highly restrictive IID and IIA conditions, provide significantly greater explanatory power than standard logit. Mixed logit and latent class MNL models exhibited the highest overall predictive accuracy on a holdout sample, while the standard logit model performed the worst. Moreover, the analysis of marginal effects of all models indicates that use of advanced models can lead to more insightful and behaviourally meaningful interpretations of the role and influence of explanatory variables and parameter estimates in model estimation. The results of this paper have implications for the use of more optimal logit structures in future research and practice. Copyright 2007 The Authors Journal compilation (c) 2007 Blackwell Publishing Ltd.

Advances in Credit Risk Modelling and Corporate Bankruptcy Prediction

The field of credit risk and corporate bankruptcy prediction has gained considerable momentum following the collapse of many large corporations around the world, and more recently through the sub-prime scandal in the United States. This book provides a thorough compendium of the different modelling approaches available in the field, including several new techniques that extend the horizons of future research and practice. Topics covered include probit models (in particular bivariate probit modelling), advanced logistic regression models (in particular mixed logit, nested logit and latent class models), survival analysis models, non-parametric techniques (particularly neural networks and recursive partitioning models), structural models and reduced form (intensity) modelling. Models and techniques are illustrated with empirical examples and are accompanied by a careful explanation of model derivation issues. This practical and empirically-based approach makes the book an ideal resource for all those concerned with credit risk and corporate bankruptcy, including academics, practitioners and regulators.

The association between the investment opportunity set and corporate financing and dividend decisions: some Australian evidence

Summarizes previous research on the investment opportunity set (IOS) using price‐based and investment‐based proxies and variance measures; and develops hypotheses on the relationship between IOS, debt/equity ratios and dividend policies. Tests them on 1990‐1998 data from listed Australian companies and explains the methodology, which builds on Gover and Gover (1993) by including more recent proxy variables. Finds no significant results from low growth firms, although some high growth firms show lower debt/equity ratios and dividends. Questions the robustness of existing IOS proxies in the Australian context and calls for further research.

The relationships among broad scope MAS, managerial control, performance, and job relevant information: A concomitant analysis

Purpose – The purpose of this paper is to demonstrate that the decision facilitation and decision influencing roles for management information are concomitant in organizations, and therefore must be analysed concurrently. Design/methodology/approach – A two-stage regression procedure for testing the moderating and intervening relationships among the variables of interest was adopted. First, the interaction of individual level controls (administrative and professional) on broad scope management accounting system (MAS) was tested, then the intervening relationship between broad scope MAS, job-relevant information (JRI), and performance was analysed. Findings – Tests of the moderating relationship between the control environment, broad scope MAS, JRI, and performance revealed a significant interaction term for JRI, but not for performance. Research limitations/implications – The usual measurement and implementation errors associated with survey methodology attach to this study. Furthermore, the sample was not randomly selected, which undermines the external validity of the findings. External validity will be further compromised by the relatively small sample size. Hence, caution is recommended in generalising the results to a wider population. Practical implications – The study provides evidence on the relationship between broad scope MAS, individual level control mechanisms and the outcome variable of JRI. Originality/value – The paper seeks to use a controlled setting, established measures, and methods of analysis to suggest that when existing theory explains the empirical relationship among the variables of interest in more than one way, it may be necessary to comprehensively examine the relationships theoretically and empirically at more than one level to enable a better explanation of the linkage between theory and the empirical tests.

Market Behavior of Institutional Investors around Bankruptcy Announcements

This paper examines, using proprietary ASX data containing institutional holdings, if institutional investors exit en mass prior to announcements of financial distress. Evidence indicates that while some institutional investors exit the stock, the withdrawal is gradual, commencing approximately 115 days prior to event. This is driven by active institutional investors reacting to the release of the financially distressed companies’ last publicly released financial reports. There is no significant decline in institutional holdings before announcements; most institutional investors hold financially distressed shares through to failure. There is evidence that the lack of disclosure drives the increase in information asymmetry prior to company failure.

Analyst Recommendations, Earnings Forecasts and Corporate Bankruptcy: Recent Evidence

This study builds on recent research by Clarke, Ferris, Jayaraman, and Lee [2006]. Based on a sample of U.S. bankruptcies between 1995 and 2001, Clarke et al. failed to find evidence of a positive bias (or overoptimism) in analyst recommendations. We extend Clarke et al.s findings by including the Global Financial Crisis (GFC) period and using an international sample of large corporate bankruptcies (the combined total assets of our firm failure sample exceed US