Multiple Discriminant Analysis Model Research Articles

Mengukur Kondisi Kesehatan Keuangan Perusahaan Sub- Industri Farmasi Saat Pandemi Covid-19

This research is intended to obtain information regarding the financial health of companies in the pharmaceutical sub-industry (code F211) on the Indonesia Stock Exchange (BEI) during the period during the Covid-19 pandemic (2020-2022). To detect whether a company is experiencing financial health problems or its finances are experiencing pressure, a multiple discriminant analysis (MDA) model is used. There are 6 MDA models used, namely: Altman model, Fulmer model, Grover model, Springate model, Taffler model, and Zmijewski model. The type of research carried out includes exploratory descriptive research. The entire population was taken as a sample, namely 11 companies belonging to the pharmaceutical sub-industry (code F211). The research results show that the use of different models produces the same results, but some also produce different results. Generally, the results of analysis of a company experiencing very severe financial pressure using several models will produce the majority of the same results. Likewise, if an analysis is carried out on a company with a very healthy financial condition using several models it will produce the same results as the majority of the models used. The research results show that there are 4 companies that need to pay attention to their financial health, namely: PT Indofarma Tbk. (INAF), PT Kimia Farma Tbk. (KAEF), PT Phapros Tbk. (PEHA), and PT Pyridam Farma Tbk. (PYFA). There are 2 companies that are indicated to have very healthy financial conditions, namely: PT Kalbe Farma Tbk. (KLBF) and PT Industri Jamu and Pharmacy Sido Muncul Tbk. (SIDO).

Ensuring the longevity of WirelessHART devices in industrial automation and control systems using distinct native attribute fingerprinting

Distinct native attribute fingerprinting is considered as a means to ensure the longevity of WirelessHART communication devices used in industrial automation and control systems. The aim is for these devices to reach full life expectancy using a technical cradle-to-grave lifecycle protection strategy. The protection addressed here includes pre-deployment near-cradle counterfeit device detection using active fingerprinting and operational mid-life rogue device detection using passive fingerprinting. The counterfeit and rogue device detection rates are estimated for 56 five-class multiple discriminant analysis models. Detection demonstrations include using three non-modeled devices to complete a total of 3 × 5 × 56 = 840 individual rogue and counterfeit device identity (ID) verification demonstrations. The device ID verification process uses binary accept/reject decisions with false positive outcomes used to estimate rogue and counterfeit device detection rates. For device ID verification using active fingerprints, the demonstrated counterfeit detection rate approached 99 % using only 15-of-99 available features—an approximate 85 % dimensional-reduction. Device ID verification using passive fingerprints was more challenging and the rogue detection rate approached 94 % using 120-of-243 available features—an approximate 50 % dimensional-reduction.. Collectively, the dimensionally-reduced implementations support efficiency improvement objectives required for providing near-cradle counterfeit device and mid-life operational rogue device detection in critical industrial automation and control systems.

Accounting‐based variables as an early warning indicator of financial distress in crisis and non‐crisis periods

AbstractFinancial integration in the Association of Southeast Asian Nations (ASEAN) region is a key focus of the ASEAN Economic Community. Whereas many studies focus on modelling corporate default, this paper identifies early warning indicators of financial distress before a default, using multiple discriminant analysis (MDA) models with a sample of listed and delisted companies in the ASEAN region. The analysis examines 720 companies in 10 different industries across six ASEAN countries from 1997 to 2016. The study constructs individual models for each country as well as an overall model for the entire region, using both in‐sample and out‐of‐sample approaches. This overall model could be useful for an integrated banking system. To ensure robustness, the study also separately examines the predictive performance of the MDA models across different economic crises: the Asian financial crisis (AFC) from 1997 to 2000, the global financial crisis (GFC) from 2007 to 2009 and their pre‐ and post‐crisis periods. We find that profitability ratios are the best indicators of financial distress in the ASEAN region, followed by liquidity and leverage ratios. In addition, our findings reveal common indicators that can be used to predict financial distress across ASEAN countries. The single model performs reasonably well in predicting financial distress 1 year ahead. In addition, the model is extended to incorporate a market‐based indicator into the MDA models, the distance to default. However, the inclusion of this indicator does not significantly improve the accuracy of the models in predicting financial distress at listed firms in the ASEAN region.

Enterprise Financial Risk Analysis Based on Improved Model C-Means Clustering Algorithm

As a provider of loans to SMEs, banks should prudently examine loan risks while ensuring that they provide loans to SMEs from the perspective of cooperating with policy implementation and controlling their own risks. The existing loan risk measurement tools include multiple discriminant analysis models, multiple regression models, and machine learning methods. Most machine learning methods have higher prediction accuracy than traditional models when using historical data for calculation, but the existence of problems such as overfitting seriously affects the robustness of machine learning methods. A similar method is introduced into the loan default risk prediction of SMEs, and the mean clustering method is used to preset penalty items to reduce overfitting and high accuracy to help banks effectively identify the default probability of SMEs during the loan period. This study will use the mean clustering method to iteratively train 900,000 SME credit records published by the US Small and Medium Business Administration, with 27 dimensions of data provided by Small Business Administration (SBA) to provide partial guarantees. A regression tree evaluates the data, combining the scores of multiple regression trees to produce a final prediction of the probability of credit default on the input data. The research results show that the mean clustering method can effectively improve the prediction accuracy of traditional machine learning methods and multiple linear regression in the scenario of SME loan default prediction and reduce the overfitting and black-box properties. As a supplementary loan default risk measurement tool, it can strengthen the ability of commercial banks to control the risk of loan business and can also promote the development of small- and medium-sized enterprises and the market economy to a certain extent.

Bankruptcy risk assessment of Polish listed companies using Asian multiple discriminant analysis models

Purpose: The main purpose of this article is to evaluate whether multiple discriminant analysis models developed in Malaysia to forecast the bankruptcy of Malaysian indus-trial companies and Singaporean companies can be useful in assessing the risk of bank-ruptcy of Polish listed companies. Methodology/approach: To test the efficacy of these discriminant models, the finan-cial statements of 25 bankrupt Polish companies and 25 viable companies were used. The accuracy of the classification of these enterprises into two groups (potential bank-ruptcy and companies able to survive on the market) was analyzed using the models developed to forecast the bankruptcy of Malaysian and Singaporean companies. Z-scores for each model were calculated for one year, two years, and three years prior to bankruptcy. Findings: The model developed to forecast the bankruptcy of Malaysian industrial companies correctly classified bankrupted Polish listed companies. However, it incor-rectly classified a large percentage of non-bankrupt companies. The Singaporean model accurately (it was 100% when data from one year prior to bankruptcy was used) assessed the risk of bankruptcy of failed companies. The percentage of correctly classified companies still in business was lower, but still acceptable. Research limitations/implications: It is recommended to test the classification abili-ties of the models using the financial data of Polish companies from various sectors of the economy (separately for selected industries), and in particular, utilizing financial reports of industrial companies. Originality/value: The Polish-language literature often describes bankruptcy forecast-ing methods developed in highly developed countries, such as the USA, Canada, Germany and Great Britain. Polish scientists have also developed many bankruptcy forecasting tools. By contrast, little attention has been paid to Asian bankruptcy forecasting methods. The study, the results of which are presented in this article, broadens the knowledge on this subject and makes it possible to test and evaluate the usefulness of such methods for Polish companies.

Financial Distress and It’s Prediction: A Case Study of the Textile and Garment Industry

Aims: The objective of this research is to glance at the projections of financial distress in the textile and garment sub-sectors listed on the IDX.
 Methodology: The case study method is used in this study to employ the descriptive quantitative method approach. While the IDX is the source of the case study data, the purposive sampling method was used on the financial statements of textile and garment sub-sector companies in 2019 and the first quarter of 2020. While the cross-sectional method is used for case study analysis, or by comparing the Z-score (multiple discriminant analysis) that has been performed between one company and the standard zone that has been carried out simultaneously.
 Results: This study discovered that the case study using multiple discriminant analysis models in the first quarter of 2020 shows a significant impact of Covid-19 on the financial condition of companies listed on the IDX in the textile and garment industry, with 88 percent of companies in a stress zone. This study also shows that both internal and external factors can lead to a company's demise. As a result, corporate financial management decision-making must consider the company's liquidity, debt proportion, and the efficient use of working capital.
 Implication/Applications: The findings of this study can be useful not only for researchers, but also for practitioners who are interested in financial distress cases.
 The Originality of the Study: One of the study's limitations is that the sample is still limited to the research scope, which only covers the two sectors. Furthermore, this study only employs a single model of financial distress. As a result, it is hoped that in the future, research will be conducted with various types of company sectors and using various financial distress models.

Predicting bankruptcy based on the full population of Croatian companies

This paper analyses the bankruptcy prediction based on the population of companies representative of the total business sector in Croatia. The representativity of the sample is achieved through the propensity score matching of the full population of bankrupt and similar non-bankrupt companies. The robust estimation of bankruptcy prediction is carried out through the multiple discriminant analysis (MDA) and logistic regression (logit). The results indicate high classification accuracy of both models, but more favourable performance of the logit estimation. Overall accuracy of the MDA model was 73.7%, while the overall accuracy of the logit model was 76.3%. The results serve as a bankruptcy estimation benchmark for the business sector in Croatia.

Bankruptcy risk assessment models as a tool for diagnosing the financial condition of capital market companies

Purpose: The main purpose of this article is to examine the usefulness of multiple discriminant analysis models in assessing the financial condition of individual enterprises, the state of the economy, and its sectors. The study assessed the financial situation of Polish listed companies at the end of the global economic crisis that started in 2007. Methodology/approach: Seven discriminant functions were used to assess the actual changes in the finan-cial situation of listed companies during the period of 2009–2014. In order to diagnose the end of the crisis, the period in which countries emerged from the global economic crisis was taken into account. The study covered 175 Polish companies listed on the regulated market operated by the Warsaw Stock Exchange, whose standalone financial statements were used. These companies belong to 22 sectors of the economy. It was assumed that the number of companies at risk of bankruptcy should have decreased during this period. Findings: The study showed that it is difficult to determine when the crisis ended and stopped affecting Polish listed companies. Their financial condition gradually improved during the period 2013–2014. Multiple discriminant analysis models are useful in assessing the risk of bankruptcy, but the study results show that it is safer to use several models simultaneously and to eliminate outliers. Research limitations/implications: The discriminant models used in the study are suitable for conducting research on large populations within enterprises and can be used by state and financial institutions (including banks) and authorities in Poland to facilitate the conduct of economic statistics, forecasting economic situation, etc. Originality/value: In Poland, many studies have been carried out on the usefulness of multiple discrimi-nant analysis models for the purposes of forecasting the bankruptcy of individual enterprises. However, there are few studies devoted to assessing the usefulness of the models in conducting research on large populations within enterprises (i.e., assessing the state of the economy and its sectors). This research helps to explore and fill this research gap.

Predicting Nursing Home Financial Distress Using the Altman Z-Score.

This article uses a modified Altman Z-score to predict financial distress within the nursing home industry. The modified Altman Z-score model uses multiple discriminant analysis (MDA) to examine multiple financial ratios simultaneously to assess a firm’s financial distress. This study utilized data from Medicare Cost Reports, LTCFocus, and the Area Resource File. Our sample consisted of 167 268 nursing home-year observations, or an average of 10 454 facilities per year, in the United States from 2000 through 2015. The independent financial variables, liquidity, profitability, efficiency, and net worth were entered stepwise into the MDA model. All of the financial variables, with the exception of net worth, significantly contributed to the discriminating power of the model. K-means clustering was used to classify the latent variable into 3 categorical groups: distressed, risk-of-financial distress, and healthy. These findings will provide policy makers and practitioners another tool to identify nursing homes that are at risk of financial distress.

Predicting Financial Distress among SMEs in Malaysia

Predicting financial distress among Small and Medium Enterprises (SMEs) can have a significant impact on the economy as it serves as an effective early warning signal. The study develops distress prediction models combining financial, non-financial and governance variables which were used to analyze the influence of major corporate governance characteristics, like ownership and board structures, on the likelihood of financial distress. Multiple Discriminant Analysis (MDA) model as one of the extensively documented approaches was used. The final sample for the estimation model consists of 172 companies with 50 percent non-failed cases and 50 percent failed cases for the period between 2000 to 2012. The prediction models perform relatively well especially in MDA model that incorporate governance, financial and non-financial variables, with an overall accuracy rate of 90.7 percent in the estimated sample. The accuracy rate in the holdout sample was 91.2 percent for the MDA model. This evidence shows that the models serve as efficient earlywarning signals and can thus be beneficial for monitoring and evaluation. Controlling shareholder, number of directors, and gender of managing director are found to be significant predictors of financially distressed SMEs.

Fingernail Trace Element Content in Environmentally Exposed Individuals and Its Influence on Their Cognitive Status in Ageing

The aim of this study was to identify links between potential long-term environmental exposure to mixed contamination and the neuropsychological performance of a group of adults (> 55 years old) living near an industrial facility. Fifty-eight permanent residents were evaluated with the Mini-Mental State Examination and Clinical Dementia Rating scale (CDR) tests. Human biomonitoring involved the measurement of trace elements in fingernails to determine exposure. Only fingernail Hg contents of the Estarreja group are above the mean content found for non-exposed and healthy centenaries populations, although several outlier values were observed for the majority of the elements. From the total of individuals recruited for the study, 29% obtained a 0 score on CDR indicating a normal status, 34% scored 0.5 which is suggestive of mild cognitive impairment (MCI), 22% had a cognitive performance indicative of mild dementia, and 14% fall under the categories of moderate and severe dementia. Principal component analysis and multiple discriminant analysis (MDA) were combined to identify groups of correlated variables discriminating well the cognitive functions examined by the different neuropsychological tests. The results obtained show that fingernails’ data discriminate well between CDR scores for normal and altered cognitive statuses, and further suggest that elevated levels of Mn and Zn in fingernails of the elderly residents of Estarreja are associated to a condition of severe dementia. The poor discriminating ability of the MDA model for intermediate categories of the cognitive tests (MCI, mild dementia) may be attributed either to a lack of sensitivity of the cognitive tests for the screening of these intermediate conditions or to the similar Zn and Mn contents in fingernail observed in participants falling under such categories.

A Look into the United Statess Underfunded Pension System

The public pension crisis has come under increasing scrutiny over the past decade as shifting demographic trends, harsh economic conditions and the very nature of pension funds have changed, and not for the better. Pension funds create valuable saving and investment tools for an individual's retirement. They make what seems like the impossible daunting task of saving sufficient funds for retirement completely feasible. All indications lead to these trends continuing, therefore pension plans need to adapt and reform. This paper is to address the pension crisis in the U.S. and intends to provide some recommendations for policy makers. This paper used the U.S. Census Bureau pension data for the fiscal years 2005-2014 to select a sample of 15 states. The time series data will be analyzed using the MDA (Multiple Discriminant Analysis) methodology to assess if a pension plan is bound to fail. MDA is used in the banking industry as a method to predict financial distress or default of bank loans. Once the regression line is determined, it can be utilized to estimate the probability of default. This methodology will be used to determine financial health of public pensions selected in the sample. The Multiple Discriminant Analysis model can be utilized to run a stress test on the public pension plans of those states selected in the sample. The Multiple Discriminant Analysis will enable public pensions and policy makers to somewhat predict the viability of their pensions. The contribution of this paper will be providing pre-warning signals and some policy recommendations for local governments to sustain their pension systems.

A new classification approach for prediction of flyrock throw in surface mines

A novel classification approach was proposed for prediction of flyrock throw distance based on the site measurements performed in a sandstone quarry. The classification approach was created by using multiple discriminant analysis (MDA) technique. The input parameters of discriminant analysis are blast design parameters and a rock mass parameter. The grouping stage of classification was performed considering a well-known blast regulation for flyrock. Additionally, multiple regression analysis technique applied to blast data to create a flyrock prediction equation. By this way, the capability and differences of the classification approach were investigated. Model validation was performed on the test blasts. MDA model successfully estimated the flyrock throw distance. Unlike the classical prediction models, the MDA model predicts severity of flyrock throws instead of a numerical value. MDA model can be easily practiced by the created territorial map. The model does not require any specific software or training for usage and suitable for practical applications at mines.

Ocena zdolności do kontynuowania działalności polskich spółek giełdowych w czasie kryzysu za pomocą modeli wielowymiarowej analizy dyskryminacyjnej

Ocena zdolności do kontynuowania działalności polskich spółek giełdowych w czasie kryzysu za pomocą modeli wielowymiarowej analizy dyskryminacyjnej The Global Financial Crisis, which began in 2007, had a huge influence on the situation on world stock markets. The behavior of investors is often affected by various factors which can impact their investment decisions. As they do not always act rationally, have a tendency to overreact and cannot remove all emo-tional components from their decision-making process, it may be difficult to explain their behavior and investment decisions during the crisis, especially those concerning the sale of shares. The huge drop in share prices on world stock markets was visible in the early stages of the crisis, but it probably was not justified by actual deterioration of the financial situation of listed companies. The Global Financial Crisis triggered a wave of panic selling of shares on the Warsaw Stock Exchange (WSE). As the fluctuations in share prices do not always reflect the real economic situation of the companies, it is worth to examine whether the financial standing of companies listed on WSE actually deteriorated and whether the number of companies facing bankruptcy grew during the 2007 Crisis. The main purpose of this article is to evalu-ate the influence of the recent financial crisis on the financial situation and performance of Polish listed companies. Eight multiple discriminant analysis models were utilized to evaluate the real changes in the financial situation of Polish listed companies during the crisis (years 2006-2011). The aforementioned models enable prediction of corporate bankruptcy and measurement of financial health of companies. Theoretically, the number of companies facing bankruptcy should increase in time of crisis. As many as 175 joint-stock companies listed on the regulated market of the WSE were covered by the study. Their financial data were extracted from the Notoria Serwis database, which is available on the University of Lodz Library’s website.

Hospitality Bankruptcy in United States of America: A Multiple Discriminant Analysis-Logit Model Comparison

ABSTRACTThis study examines bankruptcy prediction of hospitality firms within U.S. equity markets. The article investigates whether the Logit model or the Multiple Discriminant Analysis (MDA) accurately predict bankruptcy, specifically it attempts to investigate how accurate Logit and MDA models are. Various key financial variables were utilized as predictors and contrasting samples of both bankrupt and non-bankrupt firms for the period 1992–2010 were used. In this analysis Statistical software SPSS 20 was utilized for the analysis. Results show that for the period 1992–2010, the MDA model outperformed the Logit model for overall bankruptcy prediction. Theoretical and practical implications were offered based on the results. The study is critical given the significant number of hospitality enterprises being intensely impacted by the recent economic downturn. Consequently, the hospitality industry in United States demands higher degree of accuracy from bankruptcy prediction models to forecast economic failure.

COMBINING B&B-BASED HYBRID FEATURE SELECTION AND THE IMBALANCE-ORIENTED MULTIPLE-CLASSIFIER ENSEMBLE FOR IMBALANCED CREDIT RISK ASSESSMENT

An ideal model for credit risk assessment is supposed to select important features and process imbalanced data sets in an effective manner. This paper proposes an integrated method that combines B&B (branch and bound)-based hybrid feature selection (BBHFS) with the imbalanceoriented multiple-classifier ensemble (IOMCE) for imbalanced credit risk assessment and uses the support vector machine (SVM) and the multiple discriminant analysis (MDA) as the base predictor. BBHFS is a hybrid feature selection method that integrates the t-test and B&B with the k-fold crossvalidation method to search for a satisfactory feature subset. The IOMCE divides majority samples into several subsets and then combines them with minority samples to construct several training sets for constructing a multiple-classifier ensemble model. We conduct main experiments using a 1:3 imbalanced corporate credit risk data set with continuous features and extended experiments using a 1:5 imbalanced data set with continuous features and a 1:3 imbalanced data set with discrete and nominal features. We combine no feature selection and five feature selection methods (the pure B&B, the factor analysis, the pure t-test, t-test & correlation analysis, and BBHFS) with single-classifier and the IOMCE to construct SVM and MDA models for an empirical comparison. When all features are continuous, the BBHFS-IOMCE method generally outperforms all the other methods. More specifically, BBHFS provides more stable and satisfactory results than the other feature selection methods, and compared with single-classifier models, IOMCE models can significantly enhance the recognition rate for minority samples while incurring a small reduction in the recognition rate for majority samples and maintaining an acceptable overall accuracy. When the features are almost discrete or nominal, the IOMCE method retains its ability to deal with an imbalanced data set, although the five feature selection methods have no significant advantages over no feature selection. This suggests that BBHFS is effective in retaining useful information when reducing the dimensionality of continuous features and that the BBHFS-IOMCE method is an important tool for imbalanced credit risk assessment.

Small Medium Enterprises Insurance Credit Risk Management

Thai SMEs have played a vital role in the Thailand economy. However, SME has structure of short term financial characteristics as they depend mostly on short term loan. Thus, we have to be aware of financial distress of SMEs. The study has examined empirical evidence from Thailand insurance industries to identify differences between financial profiles of financially distressed SMEs and non-financially distressed SMEs. We have developed and tested a multiple discriminant analysis model to distinguish between financially distressed SMEs and non-financially distressed SMEs using three categories of financial ratios: liquidity, leverage and profitability. The results show that distressed firms SMEs had lower liquidity, higher leverage and lower profitability ratios than non-financially distressed SMEs. Study tested the statistical significance of the differences between financially distressed SMEs and non-financially distressed SMEs. The financial ratios of distressed firms were taken into the analysis to develop the prediction model. Keywords: SMEs, financially distressed SMEs, non-financially distressed SMEs, liquidity, leverage, and profitability

Assessing Credit Default using Logistic Regression and Multiple Discriminant Analysis:Empirical Evidence from Bosnia and Herzegovina

This article has an aim to assess credit default prediction on the banking market in Bosnia and Herzegovina nationwide as well as on its constitutional entities (Federation of Bosnia and Herzegovina and Republika Srpska). Ability to classify companies info different predefined groups or finding an appropriate tool which would replace human assessment in classifying companies into good and bad buckets has been one of the main interests on risk management researchers for a long time. We investigated the possibility and accuracy of default prediction using traditional statistical methods logistic regression (logit) and multiple discriminant analysis (MDA) and compared their predictive abilities. The results show that the created models have high predictive ability. For logit models, some variables are more influential on the default prediction than the others. Return on assets (ROA) is statistically significant in all four periods prior to default, having very high regression coefficients, or high impact on the model's ability to predict default. Similar results are obtained for MDA models. It is also found that predictive ability differs between logistic regression and multiple discriminant analysis.

Comparing Logit, Probit and Multiple Discriminant Analysis Models in Predicting Bankruptcy of Companies

Experimental studies across the world show that one of investors and other activists’ expectations in capital market is their ability in predicting the status of activity consistency or the bankruptcy of the companies. For this purpose, this study tries to predict the bankruptcy likelihood of manufacturing firms and comparing the predictability power of 3 methods of Logit, Probit, and Multiple-discriminant analysis. Statistical population of the study included all manufacturing firms listed in Stock Exchange of Tehran since 2000-2010. The results showed that the predictability of Logit, Probit, and Multiple-discriminant models has been81, 80, and % 70 respectively. The results also showed the higher predictability and accuracy of Logit model and also Multiple discriminant analysis methods identify bankrupt firms with high accuracy in the level logit model.

Application of multiple discriminant analysis (MDA) as a credit scoring and risk assessment model

PurposeThe purpose of this research is to investigate whether inclusion of risk assessment variables in the multiple discriminant analysis (MDA) model improved the banks ability in making correct customer classification, predict firm's performance and credit risk assessment.Design/methodology/approachThe paper reviews literature on the application of financial distress and credit scoring methods, and the use of risk assessment variables in classification models. The study used a sample of 56 performing and non‐performing assets (NPA) of a privatized commercial bank in Tanzania. Financial ratios were used as independent variables for building the MDA model with a variation of five MDA models. Different statistical tests for normality, equality of covariance, goodness of fit and multi‐colinearity were performed. Using the estimation and validation samples, test results showed that the MDA base model had a higher level of predictability hence classifying correctly the performing and NPA with a correctness of 92.9 and 96.4 percent, respectively. Lagging the classification two years, the results showed that the model could predict correctly two years in advance. When MDA was used as a risk assessment model, it showed improved correct customer classification and credit risk assessment.FindingsThe findings confirmed financial ratios as good classification and predictor variables of firm's performance. If the bank had used the MDA for classifying and evaluating its customers, the probability of failure could have been known two years before actual failure, and the misclassification costs could have been calculated objectively. In this way, the bank could have reduced its non‐performing loans and its credit risk exposure.Research limitations/implicationsThe valiadation sample used in the study was smaller compared to the estimation sample. MDA works better as a credit scoring method in the banking environment two years before and after failure. The study was done on the current financial crisis of 2009.Practical implicationsUse of MDA helps banks to determine objectively the misclassification costs and its expected misclassification errors plus determining the provisions for bad debts. Banks could have reduced the non‐performing loans and their credit risks exposure if they had used the MDA method in the loan‐evaluation and classification process. The study has proved that quantitative credit scoring models improve management decision making as compared to subjective assessment methods. For improved credit and risk assessment, a combination of both qualitative and quantitave methods should be considered.Originality/valueThe findings have shown that using the MDA, commercial banks could have improved their objective decision making by correctly classifying the credit worthiness of a customer, predicting firm's future performance as well as assessing their credit risk. It has also shown that other than financial variables, inclusion of stability measures improves management decision making and objective provisioning of bad debts. The recent financial crisis emphasizes the need for developing objective credit scoring methods and instituting prudent risk assessment culture to limit the extent and potential of failure.