Search results

This study updates the literature review of Jones (1987) published in this journal. The study pays particular attention to two important themes that have shaped the field over the past 35 years: (1) the development of a range of innovative new statistical learning methods, particularly advanced machine learning methods such as stochastic gradient boosting, adaptive boosting, random forests and deep learning, and (2) the emergence of a wide variety of bankruptcy predictor variables extending beyond traditional financial ratios, including market-based variables, earnings management proxies, auditor going concern opinions (GCOs) and corporate governance attributes. Several directions for future research are discussed.

This study provides a systematic review of the corporate failure literature over the past 35 years with a particular focus on the emergence of new statistical learning methodologies and predictor variables. This synthesis of the literature evaluates the strength and limitations of different modelling approaches under different circumstances and provides an overall evaluation the relative contribution of alternative predictor variables. The study aims to provide a transparent, reproducible and interpretable review of the literature. The literature review also takes a theme-centric rather than author-centric approach and focuses on structured themes that have dominated the literature since 1987.

There are several major findings of this study. First, advanced machine learning methods appear to have the most promise for future firm failure research. Not only do these methods predict significantly better than conventional models, but they also possess many appealing statistical properties. Second, there are now a much wider range of variables being used to model and predict firm failure. However, the literature needs to be interpreted with some caution given the many mixed findings. Finally, there are still a number of unresolved methodological issues arising from the Jones (1987) study that still requiring research attention.

The study explains the connections and derivations between a wide range of firm failure models, from simpler linear models to advanced machine learning methods such as gradient boosting, random forests, adaptive boosting and deep learning. The paper highlights the most promising models for future research, particularly in terms of their predictive power, underlying statistical properties and issues of practical implementation. The study also draws together an extensive literature on alternative predictor variables and provides insights into the role and behaviour of alternative predictor variables in firm failure research.

This study examines the relationship between managerial overconfidence and corporate financing decisions by constructing proxies for managerial overconfidence based on the track records of earnings forecasts in Japanese listed firms. We find that managers have the stable tendency to forecast overly upward earnings compared to actual ones and that their upward bias decreases the probability of issuing equity in the public market by about 4.7 percent per one standard error, which economically has the strongest impact on financing decisions. This tendency is observed when we employ alternative measures for managerial overconfidence and other model specifications. However, in private placements, the choice to offer equity is not always avoided by managers. This implies that managers place private equity with the expectation of the certification effect

In this chapter, we describe how random utility maximization (RUM) discrete choice models are used to estimate the demand for commodity attributes in quality-differentiated goods. After presenting a conceptual overview, we focus specifically on the conditional logit model. We examine technical issues related to specification, interpretation, estimation, and policy use. We also discuss identification strategies for estimating the role of price and non-price attributes in preferences when product attributes are incompletely observed. We illustrate these concepts via a stylized application to new car purchases, in which our objective is to measure preferences for fuel economy.

With this study we investigate the preferences of an important category of consumers of organic products (regular consumers of organic food or RCOFs) allowing for preference heterogeneity. A survey instrument was developed to elicit preferences for important qualitative and quantitative attributes of extra virgin olive oil. The survey was administered via questionnaire to a random sample of 198 RCOFs in organic food stores of Naples, Italy. The choice task was organised around a fractional factorial main effects orthogonal design. Each respondent made eight choices to rank‐order nine product profiles in terms of their individual preference. Product attributes included price, origin of production, type of certification and visual appearance. Interestingly, the set of observed responses appears to display significant preference heterogeneity for origin of production and price. Once heterogeneity and correlation among repeated choice by the same respondent are accounted for by means of random‐parameter panel logit models, the fit increases dramatically with respect to the more restrictive fixed‐parameter logit models. Results also suggest that price plays an important role as quality proxy, while visual appearance is not significant in preference modelling and the type of certification programme has a fixed effect.

Patients and health professionals often make decisions which involve a choice between discrete alternatives. This chapter reviews the econometric methods which have been developed for modelling discrete choices and their application in the health economics literature. We start by reviewing the multinomial and mixed logit models and then consider issues such as scale heterogeneity, estimation in willingness to pay space and attribute non-attendance.

Improved indigenous chicken is considered a sustainable agricultural practice with social, economic and environmental indicators. Therefore, the analysis of the choice of market channels is of considerable importance to farmers with reference to improved livelihoods and poverty alleviation in developing countries. The purpose of this study is to investigate the factors that influence market channel choices among improved indigenous chicken farmers in Baringo County and to rank the determinants according to their level of importance in influencing farmer's choice of marketing channels.

A multistage sampling technique was employed to collect data from 209 households for the study conducted between April and July 2019, out of which, 198 useful responses were obtained. Multinomial logit regression and neural network models were used to analyze the factors influencing market channel choice based on socioeconomic, demographic and farm characteristics.

It was established that group membership, education, market distance, transport costs, farm size, cost of information and bargain costs were statistically significant in the choice of market channels (wholesaler, brokers, processors and supermarkets). With the direct consumer as the base market choice. The cost of transport had the highest normalized importance in the prediction of a farmer's selection of market channels for both radial basis function (RBF) and multilayer perceptron (MLP) neural networks. However, flock attributes and age of household head had the least normalized importance in MLP and RBF, respectively.

Due to the insufficiency of resources and time, this study only focused on a small part of the country (Baringo County). However, improved indigenous chicken farming is widely practiced in Kenya. Further studies can be carried out in other counties to validate the results of this study.

The outcome can be used in policy implementation involving improved indigenous chicken production in Kenya.

This study suggests the methods aimed at enhancing poultry sector in other counties in Kenya as well as other developing countries.

This paper aims at investigating the link between inter-organizational relationships and innovation, discussing whether the probability to detect a greater innovation propensity of organizations increases with or without collaborative partnerships.

Drawing on secondary data provided by the Italian Institute of Statistics (ISTAT) about a representative sample of 8,967 Italian firms, three multinomial logit models and four logit models have been estimated, in an attempt to examine the effects of inter-organizational relationships on different types of organizational innovation.

A positive and statistically significant relationship between inter-organizational relationships and organizational innovation emerged from all the models which were arranged for the purpose of this study. Several categories of partners, including suppliers, universities and firms belonging to the same holding group, were found to be more effective in fostering the probability of organizational innovation. Interestingly, geographical proximity did not seem to influence the organizational propensity to innovate.

Even though further developments are needed to disentangle the complex link between inter-organizational relationships and organizational innovation, the former are likely to positively affect the innovation ability of organizations. From this point of view, it could be argued that partners perform as catalysts, which boost the knowledge creation process underlying the emergence of organizational innovation.

This is one of the first attempts to exploit the potential of multinomial logit models and logit models to investigate the effects of inter-organizational relationships on the propensity of organizations to innovate.

This paper aims to investigate the factors associated with the voluntary disclosure of integrated reporting (IR) in Brazil of the companies listed on the stock exchange – Brasil, Bolsa, Balcão. The cultural dimensions of a nation reflect different priorities in accounting practices. The Brazilian case, therefore, becomes significant, as Brazil is increasingly important in world markets.

As an explanatory econometric model, multinomial logistic regression was used (Y = 2 to describe the probability of the IR disclosure; Y = 1 to describe the occurrence of reports with practices similar to the IR and Y = 0 to describe the occurrence of non-disclosure of non-financial reports). Applied to panel data with random effects (chosen for best performance) in the period from 2016 to 2019.

Reveals the positive association of the company’s profitability and market-to-book with the probability of the IR disclosure. Regarding the board composition, it is suggested that size does not make a difference, with the greater participation of women and independence of directors associated with better probabilities of adopting the IR in Brazil.

This work is the first to characterize the Brazilian reality of voluntary disclosure, specifying the implementation of IR, compared to publishing reports similar to the IR and not reporting structured non-financial statements. It can also be considered the first study on the relevance of the board structure in the disclosure of IR by Brazilian companies. Finally, it contributes to the literature on IR adoption, bringing practical results in understanding the most favorable conditions in which the IR framework will be fully implemented.