Search results

The purpose of this paper was exploring and comparing different deterioration models based on statistical and machine learning approaches. These models were chosen from their successful results in other case studies. The deterioration models were developing considering two scenarios: (i) only the age as covariate (Scenario 1); and (ii) the age together with other available sewer characteristics as covariates (Scenario 2). Both were evaluated to achieve two different management objectives related to the prediction of the critical condition of sewers: at the network and the sewer levels.

Six statistical and machine learning methods [logistic regression (LR), random forest (RF), multinomial logistic regression, ordinal logistic regression, linear discriminant analysis and support vector machine] were explored considering two kinds of predictor variables (independent variables in the model). The main propose of these models was predicting the structural condition at network and pipe level evaluated from deviation analysis and performance curve techniques. Further, the deterioration models were exploring for two case studies: the sewer systems of Bogota and Medellin. These case studies were considered because of both counts with their own assessment standards and low inspection rate.

The results indicate that LR models for both case studies show higher prediction capacity under Scenario 1 (considering only the age) for the management objective related to the network, such as annual budget plans; and RF shows the highest success percentage of sewers in critical condition (sewer level) considering Scenario 2 for both case studies.

There is not a deterioration method whose predictions are adaptable for achieving different management objectives; it is important to explore different approaches to find which one could support a sewer asset management objective for a specific case study.

The originality of this paper consists of there is not a paper in which the prediction of several statistical and machine learning-based deterioration models has been compared for case studies with different local assessment standard. The above to find which is adaptable for each one and which model is adaptable for each management objective.

The purpose of this study is to demonstrate that the internal ratings-based (IRB) approach provides more effective risk discrimination than the standardized approach when calculating regulatory capital for retail credit risk exposures.

The author uses four retail credit data sets to compare regulatory capital appropriation using the IRB approach and the standardized approach. The author follows the regulatory capital calculation method recommended under Basel III. For the IRB approach, the author uses a logistic regression to determine the probability of default.

The results suggest that the IRB approach provides more effective risk discrimination across individual exposures, which allows more regulatory capital to be held against riskier exposures and less regulatory capital to be held against less risky exposures. The author further argues that the Basel III output floor, as presently constructed, may disincentivize the use of the IRB approach and further diminish the value of secured lending under the IRB approach. To address this issue, the author offers two simple adjustments to the current design of the output floor.

While studies have argued the idea of risk-sensitive regulatory capital, the author has not observed any research that empirically compares the risk-sensitivity of regulatory capital across retail credit exposures, which makes up a significant portion of many banks’ credit exposures. This study also highlights what appears to be a major point of concern for the output floor, which is set to be phased in starting January 2022. This is of particular value because this point has not appeared to receive any attention in the literature thus far.

The purpose of this paper is to develop a model for assessing the audit evidence of the going-concern (GC) assumptions underlying the preparation of financial statements.

This research analyses 678 audit opinions of Italian listed firms from 2007 to 2016 and uses a multiple linear discriminant analysis to create a GC score, which includes variables suggested by the international standards on auditing (ISA) 570 and by literature on GC.

The model provides three cut-off scores which can orient auditors towards issuing the most appropriate GC audit opinions (unmodified opinion, unmodified opinion, which includes emphases of matter, qualified opinion or disclaimer of opinion).

The development of the model is mainly based on public data and does not assess confidential information that is not disclosed in audit opinions.

This model can enable auditors to identify the most appropriate GC opinion and align auditor’s opinions in similar circumstances, thereby reducing their reliance on discretion and increasing the reliability of their judgement with a higher degree of accuracy. Moreover, this research lists additional events or conditions that may individually or collectively cast significant doubt on GC assumptions.

This study goes beyond the traditional decision-making process, apparently binary in nature, between “continuity” and “failure” or between “unmodified” and “modified” opinions. It is conceived to detect the different degrees of uncertainty that affect GC evaluations to orient auditors’ professional judgements.

The purpose of this paper is to discuss and assess the structural characteristics (conceptual utility) of the most popular classification and predictive techniques employed in customer relationship management and customer scoring and to evaluate their classification and predictive precision.

A sample of customers' credit rating and socio‐demographic profiles are employed to evaluate the analytic and classification properties of discriminant analysis, binary logistic regression, artificial neural networks, C5 algorithm, and regression trees employing Chi‐squared Automatic Interaction Detector (CHAID).

With regards to interpretability and the conceptual utility of the parameters generated by the five techniques, logistic regression provides easily interpretable parameters through its logit. The logits can be interpreted in the same way as regression slopes. In addition, the logits can be converted to odds providing a common sense evaluation of the relative importance of each independent variable. Finally, the technique provides robust statistical tests to evaluate the model parameters. Finally, both CHAID and the C5 algorithm provide visual tools (regression tree) and semantic rules (rule set for classification) to facilitate the interpretation of the model parameters. These can be highly desirable properties when the researcher attempts to explain the conceptual and operational foundations of the model.

Most treatments of complex classification procedures have been undertaken idiosyncratically, that is, evaluating only one technique. This paper evaluates and compares the conceptual utility and predictive precision of five different classification techniques on a moderate sample size and provides clear guidelines in technique selection when undertaking customer scoring and classification.

Like all other natural fibers, the physical properties of cotton also vary owing to changes in the related genetic and environmental factors, which ultimately affect both the mechanics involved in yarn spinning and the quality of the yarn produced. However, information is lacking about the degree of influence that those properties impart on the spinnability of cotton fiber and the strength of the final yarn. This paper aims to discuss this issue.

This paper proposes the application of discriminant analysis as a multivariate regression tool to develop the causal relationships between six cotton fiber properties, i.e. fiber strength (FS), fiber fineness (FF), upper half mean length (UHML), uniformity index (UI), reflectance degree and yellowness and spinning consistency index (SCI) and yarn strength (YS) along with the determination of the respective contributive roles of those fiber properties on the considered dependent variables.

Based on the developed discriminant function, it can be revealed that FS, UI, FF and reflectance degree are responsible for higher YS. On the other hand, with increasing values of UHML and fiber yellowness, YS would tend to decrease. Similarly, SCI would increase with higher values of FS, UHML, UI and reflectance degree, and its value would decrease with increasing FF and yellowness.

The discriminant functions can effectively envisage the contributive role of each of the considered cotton fiber properties on SCI and YS. The discriminant analysis can also be adopted as an efficient tool for investigating the effects of various physical properties of other natural fibers on the corresponding yarn characteristics.

In recent years, a sizeable amount of research in finance and accounting has been devoted to the issue of bond rating and bond rating changes. A major thrust of these research efforts was to develop and test some prediction‐based models using mainly financial ratios and their trends. This paper tests the ability of statistical decomposition analysis of financial statements to predict bond rating changes. The results show that the decomposition analysis almost does not beat the a priori probability model and is no better than multiple discriminant analysis using simple financial ratios. One important piece of information for participants in debt markets is the assessment of the relative risk associated with a particular bond issue, commonly known as bond ratings. These ratings, however, are not usually fixed for the life of the issues. From time to time, the rating agencies review their ratings of the outstanding bond issues and make changes to these ratings (either upward or downward) when needed. Over the years, researchers have attempted to develop and test some prediction based models in order to predict bond ratings or bond rating changes. These prediction models have employed some variables that are assumed to reflect the rating agency decision‐making activities. Although the rating process is complicated and based mainly on judgmental considerations, Hawkins, Brown and Campbell (1983, p. 95) reported that the academic research strongly suggests that a reliable estimate of a potential bond rating or rating change can be determined by a few key financial ratios. Information theory decomposition measures have received in recent years considerable attention as a potential tool for predicting corporate events, namely corporate bankruptcy (e.g., Lev 1970; Moyer 1977; Walker, Stowe and Moriarity 1979; Booth 1983). The underlying proposition in these studies is that corporate failure, as an event, is expected to be preceded by significant changes in the company's assets and liabilities structure. Although the event of bond rating changes is different from the bankruptcy event in terms of consequences, one can still propose that a bond rating change, as a corporate event, is also expected to be preceded by some significant changes in the company's assets and liabilities structure. Therefore, the decomposition analysis may have a predictive ability in the case of bond rating changes. The purpose of this paper is to empirically test and compare the classification and predictive accuracy of the decomposition analysis with the performance of a multiple discriminant model that uses financial ratios and their trends in the context of bond rating changes.

Prior studies have demonstrated that simple linear discriminant models can be highly successful in identifying financially distressed companies, and therefore useful in predicting corporate failures. Such models have been shown to be both industry and country specific even though their variable selection has been narrow. These models have remained incredibly robust over time despite variations in the definition of the ‘distressed’ state employed for modelling purposes. This paper extends such analysis to the main and second boards of the Kuala Lumpur Stock Exchange (KLSE) in Malaysia, with particular reference to their designation of PN4 companies (those classified as ‘distressed’ in accordance with Practice Note No. 4 introduced in February 2001). The findings of the study show that a single discriminant model has high classificatory power for both boards of the KLSE, and that the optimum model comprises financial ratio variables common to other published models. Previous findings are therefore shown to be substantially generalisable to a new environment and to a different definition of distress.

The purpose of this paper is to perform a comparative study of prediction performances of an artificial neutral network (ANN) model against a linear prediction model like a linear discriminant analysis (LDA) with regards to forecasting corporate credit ratings from financial statement data.

The ANN model used in the study is a fully connected back‐propagation model with three layers of neurons. The paper uses a comparative approach whereby two prediction models – one based on ANN and the other based on LDA are developed using identically partitioned data set.

The study found that the ANN model comprehensively outperformed the LDA model in both training and test partitions of the data set. While the LDA model may have been hindered by omitted variables; this actually lends further credence to the ANN model showing that the latter is more robust in dealing with missing data.

A possible drawback in the model implementation probably lies in the selection of the various accounting ratios. Perhaps future replications of this study should look more carefully at choosing the ratios after duly addressing the problems of collinearity and duplications more rigorously.

The findings of this study imply that since ANN models can better deal with complex data sets and do not require restraining assumptions like linearity and normality, it may be overall a better approach in corporate credit rating forecasts that uses large financial data sets.

This study brings out the effectiveness of non‐linear pattern learning models as compared to linear ones in forecasts of financial solvency. This goes on to further highlight the practical importance of the new breed of computational tools available to techno‐savvy financial analysts and also to the providers of corporate credit.

The purpose of this paper is to appraise methodological rigor in the application of discriminant analysis (DA) in export-focused research and to offer guidelines for future studies.

The sample includes 89 empirical peer-reviewed studies, comprising 102 models published over the period 1979-2014. Content analysis and vote counting are used to evaluate each of these studies.

This review highlights major flaws in the application of DA in export research. The shortcomings are self-evident particularly concerning suitability of DA for research context, completeness in the reporting of descriptive results, and validity and reliability of predictive results.

The study takes the position that the lack of methodological rigor may be undermining the eminence of knowledge in exporting, and this has extensive implications for both researchers and practitioners.

This review outlines steps to assess methodological rigor associated with DA and offers guidelines for scholars seeking to enhance rigor in future research.