Abstract
The present study is mainly devoted to the bankruptcy prediction models and their ability to assess a bankruptcy probability for oil drilling and exploration sector of Indian. The study puts an effort to determine the financial health of 12 selected companies from this sector of India for a period of 5 years. These companies serve the backbone of many other industries such as transport industry, manufacturing industry, automobile industry and so on of the Indian economy. The study has taken the reference of Altman’s Z-score model, where ratios such as working capital to total asset, retained earnings to total asset, earnings before interest and tax to total assets, market value of equity to book value of debt and sales to total assets have been taken. The discriminant analysis is conducted to validate the outcomes of Altman’s model to predict group membership and to forecast the overall industry condition. The study reveals that 75 per cent of the companies are in financially healthy zone. The results indicate that working capital/total assets can very well explain the Z-score. The research on financial health using Altman’s score is very limited in Indian context. Therefore, this study will add value to the existing body of literature for financial risk.
Introduction
Every firm is operating in the modern cut-throat competitive world for earning profit. Profitability reveals the financial soundness as well as the profit position of the industry. The profitability aspect could be visualized by the company’s financial statement analysis (Agatha, 2007). A thorough analysis of financial statement helps to know the trend and changes in financial position of the companies over the periods of time. This information became very useful for the stakeholders of the company. Stakeholders will remain interested to know both short- and long-term financial soundness of the company (Gumparthi & Manickavasagam, 2010). In this respect, financial distress refers to the situation when the company faces difficulty in paying off its obligation to the creditors. In long term, this financial distress might lead to bankruptcy to the organization. The role of discriminant analysis as developed by Altman is a reliable and proven statistical tool to calculate the financial distress of the company (Ante & Ana, 2013). The present study focuses on determining the financial distress of selected companies in ‘oil drilling and exploration’ sector of latest 5 years.
The Scenario of Global Oil and Gas Sector
The traditional source of oil and gas industry has been changing and some new areas have been explored such as shale gas in the US, oil sands in Canada, coal seam gas in Australia, and deep-water offshore wells in Brazil, West Africa and Asia. The changes are due to technological changes, environmental changes and sustainability factors. Since the beginning, the oil industry gives a response to challenges that have implications on global economy and environment. Earlier, ‘oil industry’ controlled the transport sector which in turn controlled all other industries (British Petroleum [BP], 2018). However, since 2005, the government policies limited carbon emissions that led to the various researches on non-conventional energy. Now oil industry has no deep impact on transport industry. Consequently, the role of OPEC gradually changed. Competition from outside oil industry is a real and present threat to demand oil. Research work is going on to develop plug-in hybrids and battery electric vehicles (BEVs). This might transform the demand for liquid fuel, the shape of the automotive industry, the infrastructure of a country and overall economy (BP, 2018).
Oil and Gas Industry in India
‘The Indian oil drilling and exploration’ sector is one of the six core sectors of the Indian economy. It has a huge impact on the other spheres of the economy. This sector witnesses huge growth and it will accelerate in future; consequently, it encourages all-round growth and development of the economy. The strategic decisions like the formation of New Exploration Licensing Policy (NELP) in 1997–1998 have become imperative in attracting both foreign and domestic investment in India. The government also decides to allow 100 per cent foreign direct investment (FDI) in petroleum products. One of the biggest FDI deals happened in India during 2011 as British Petroleum formed a partnership with Reliance Petroleum. India’s domestic crude oil production of 36.95 million tons in 2015–2016 barely met 20 per cent of its oil needs (BP, 2018). In 2016–2017, India consumed 193.745 MMT of petroleum products, while the consumption stood at 184.674 MMT during 2015–2016. In 2017–2018, up to September, the figure stood at 96.82 MMT. India has always been an import-dependent nation in the ‘oil and natural gas’ (O&NG) sector (BP, 2018).
Oil import rises sharply by 28 per cent to US$9.29 billion in 2017. It is the third largest oil-consuming nation in the world. In India, the O&NG industry has huge potential and contributes over 15 per cent to India’s GDP.
Company Details of OIL and Gas Exploration Sector of India
Dolphin Offshore Enterprises (India) Limited is a holding company, which is engaged in offshore business. The company offers a range of services including vessel operations and management, electrical and instrumentation services, and offshore hook-up and commissioning. The Selan Exploration Technology Limited (SELAN) is a private sector listed company engaged in oil exploration and production since 1992. The promoters have extensive experience and domain knowledge in the field of petroleum exploration, development and production as well as in the field of geophysical data acquisition, processing and interpretation. Gujarat State Petroleum Corporation Ltd (GSPC) is a group of oil and gas exploration, production and distribution companies based in Gujarat. The Deep Industries Limited started in the year 1991, with a primary objective of catering the ever-increasing demand for ‘oil and gas industry’. Since its inception, Deep Industries Limited has been serving the industry in various segments: providing natural gas compressor services on chartered hire basis tops all wherein company commands a healthy market and reported net profit of ₹76.02 crores. Petronet LNG Limited, one of the companies in the Indian energy sector, has set up the country’s first LNG receiving and regasification terminal at Dahej, Gujarat, and another terminal at Kochi, Kerala. Indian Oil Corporation Limited (IOCL), commonly known as Indian Oil, is an Indian state-owned oil and gas company, largest commercial oil company in the country, with a net profit of 19,106 crore for the financial year 2017–2018. The Gail (India) Limited is the largest state-owned natural gas processing and distribution company in India. It has the following business segments: natural gas, liquid hydrocarbon, liquefied petroleum gas transmission, petrochemical, city gas distribution, exploration and so on. The Aban Offshore is India’s largest offshore drilling services provider to oil companies, mainly for the Oil and Natural Gas Corporation (ONGC). The group has also ventured into construction, offshore and onshore drilling, wind energy and power generation, information technology enabled services, hotels and resorts, tea plantations and marketing. Hindustan Oil Exploration Company Limited (HOEC) is involved in exploration and production of hydrocarbons, which are natural resources. Its segments include hydrocarbon and oil additives. The Cairn India, an Indian oil and gas exploration and production company, is a subsidiary of Vedanta Resources Limited. Cairn India is one of the largest independent oil and gas exploration and production companies in India. The Alphageo (India) Limited is the largest onshore integrated seismic service provider in the private sector and enjoys a market leadership in seismic survey. The company’s globally benchmarked competence comprises acquisition, processing and interpretation of seismic data for oil exploration. ONGC is an Indian multinational oil and gas public sector unit company. It produces around 70 per cent of India’s crude oil (equivalent to around 30 per cent of the country’s total demand) and around 62 per cent of its natural gas.
These 12 companies are main backbone of the Indian oil and drilling sector and contribute almost 92 per cent market capital of Indian market because of this, we have selected these 12 companies as a representative of Indian oil drilling sector and further proceed with these companies.
Analysis of Financial Indicators and Coefficients
Companies carry out the financial analysis not only to assess the company’s current financial condition but it also allows predicting its further development. In this case, analysts need to consider the list of indicators carefully that will be used for strategic planning (Bhunia, 2010).
Analysis of the level of sustainable growth of the company is a dynamic analytical framework that combines financial analysis with strategic management to explain important interrelationships between the variables of strategic planning and financial variables and to verify the adequacy of corporate growth and financial policies. This analysis allows determining the presence of the company’s existing opportunities for financial growth to establish how the company’s financial policy will influence the future and analyses the strengths and weaknesses of the company’s competitive strategies (Platt & Platt, 2006). Any activities to implement strategic programmes have their value. A necessary part of the planning and implementing the strategy is to calculate the necessary and sufficient financial resources that the company must invest (Demyanyk & Hasan, 2010).
Information for Financial Analysis
The most complete definition of the concept of financial analysis is given in the Financial and Credit Encyclopedic Dictionary, ‘Financial analysis is a set of methods for determining the property and financial position of an economic entity in the past period, as well as its opportunities for the near and long term’. The main objective of financial analysis is to determine the most effective ways to achieve profitability of the company; the main tasks are profitability analysis and enterprise risk assessment (Jayadev, 2006). The analysis of financial indicators and ratios will permit the manager to understand the competitive position of the company at the current time. However, the published reports and accounts of companies contain many figures, and the ability to read this information allows analysts to know how efficiently and effectively these companies and competitors operate. The financial ratios allow us to visualize the relationship between the sales profit and the expenses, and the main assets and the liabilities. There are some ratios which are usually used to analyse the five main aspects of the company, that are, liquidity, the ratio of equity and debt, asset turnover, profitability and market value (Gurný & Gurný, 2013). This indicator closely determines the related financial status of an organization and afterward there is a possibility that the contrast between the likelihood of the best organizations and worst organization in those more next period searches for linear combinations. Numerous financial institutions and analysts are quick to develop methods to predict the company performance by making score function. The procedure utilized is a discriminant analysis that considers numerous financial and economic aspects of the organization and the second step is to use statistical techniques for determining its significance (Mileris, 2010). Now, for more understanding of the financial distress and selecting the best way to analyse these distresses, the authors search the literatures to visualize the measurement.
Therefore, this study is motivated to understand how financial performance of oil drilling and exploration sector is affected by the financial distress. This will enable the companies of this sector to take corrective measures in due time if they find themselves in distress to avoid the shocking results as there is a quote called ‘Prevention is better than cure’. To analyse the financial performance of this sector, we have to choose a model or technique, which is widely accepted for financial distress analysis. For that reason, we have to do extensive literature review. In the next section, the literatures belonging to the financial performance analysis are discussed.
Literature Review
Most of the current literatures on financial distress depicted the importance of various ratios such as profitability, solvency and liquidity of the companies. The authors for predicting the financial distress and probable bankruptcy have used the various univariate and multivariate statistical techniques. In 1967, Beaver has taken 37 financial ratios for select US companies. There are 79 paired failed and non-failed US companies as per the authors’ selection for their study. It has been found that there was a considerable dissimilarity between the Z-score of the companies in both the segments. Some comparisons have been made to judge the predictability of the Z-score of the pairs of firms. The author has identified three ratios that are named as the total flow of cash flows, the total income of the total assets and the total assets ratio, which is the best indicator of the debt. The study also revealed that the ratios of failed firms varied significantly from those of non-failed firms and they deteriorated severely during the 5 years prior to failure.
Altman (1968) has discussed the Z-score model about 66 organizations by utilizing the multiple discriminatory analyses to the incompetent failing organizations and predicted the probability of failure. The five weighted financial proportions have been chosen, that are, the effective capital for total liability, the overall property interest and the number of earned income before taxes, the sum of the total book value of the loan and the sale of total assets. The above things are able to predict bankruptcy with 45 per cent degree of accuracy. The author also revealed that the predictive capacity of the model destabilized very sharply when the number of years preceding to the failure increased. This study is based on the estimation of the financial position of the companies cited in Malaysia’s stock exchange about the applicability and usefulness of the Altman’s model (Mohammed, Abusalah, & Ng, 2013). The authors have given importance to the Altman’s model and current ratio. This study revealed that there are financially distressed companies listed on the main board that are not classified as PN17 companies.
Grunet et al. (2004) have used non-financial variables as predictors of failure of 25 companies, which shows significant improvement in the prediction model’s accuracy. The findings clearly confirm for small- and medium-sized enterprises (SMEs) what has been found in other studies for large corporations. However, the authors believe that these results for SMEs are more important as a large part of the financial information is often limited to them. Furthermore, apart from accounting information, some non-accounting information could often be updated frequently so that financial institutions allow their credit decisions to be set at the right time. In this way, banks should definitely consider the results of this study while setting up internal systems and procedures in the management of credit risk for SMEs.
The past research and the research study by Altman and Sabato (2007) established that banks should separate small- and medium-sized firms from large corporations when they are setting their credit risk systems and strategies. The main idea of the current study focuses specifically on SME. SME segmentation models and methods are needed, but authors analyse the study on a new geographic area (UK) including about 6 million SMEs using samples. Primarily, the selected study adds non-financial information that reflects the company’s description of predictive variables of company statements and the effective risk profile of financial reporting compliance, internal audit and trade credit relations. We improve on existing models from the distressed SME literature in various ways. This research is capable of using non-financial information specifically for SMEs. We find that when this information is available, the forecast will increase the accuracy by 13 per cent. In all the models, we control macroeconomic conditions, for a 51 per cent industry sector compared to the previous year, for every company, with a failure rate associated with the industrial sector. By using existing information, the author creates a default prediction model for a large part of SME for which financial information is very limited (for example, business, professional, micro company, simple accountancy or tax reporting agency). The best known of the writers of existing literature, credit risk management solutions have never been provided for these clients.
Gumparthi and Manickavasagam (2010) have endeavoured to do the risk classification for SMEs. In their study, 140 samples have been chosen, which are classified as a performing company and non-performing company (NPC) based on their Z-score through discriminant analysis. More sub-classification was done which has added the significance level. The model was deeply stabilized by this class of attribute, which added the third feature of this model. The model has three important features, which are not available in the market, such as ratings, disparity between ratings, good and bad assets, and risk classification.
Researchers have studied the multivariate statistics system, which has confirmed that according to the financial ratios, the banks can be classified into above or below average profit groups with profitable or non-profitable and maximum high accuracy (Ante & Ana, 2013). In the study, the profitability of the bank remains an important issue. Inequality analysis of this article is done with a goal of recognizing the profitability determinants of the bank. The present study uses the multivariate statistical method. The study also confirmed that according to the financial ratios, banks could be classified into groups of profitable or non-profitable banks. The banks could also be grouped into above or below average profitability with very high precision. In this article, discriminant analysis was carried out with a goal of recognizing determinants of bank profitability.
Sustainability in bank profitability is the building block of its auto-financing process and stability. The outcome of discriminant analysis shows the high classification accuracy when the profit of the bank comes out through its survival and existence. The minimum decision on determinants on the profitability of the banks found in conventional literature has been reached. Here, the bank profitability was defined as a categorical variable and discriminant analysis was employed, which is unusual for this research problem; obtained results are consistent with theoretical explanations and empirical framework and possess statistical and economic significance.
The probability of default in risk management perspective includes evaluation of credit derivatives and creditworthiness credentials and estimation of capital adequacy estimates of banks as given by Gurný and Gurný (2013). The false assessment of the probability of the default assesses the false assessment of the risk and, consequently, the financial problems of certain companies might be encountered. In their article, the authors have described the probability of default estimation of the US commercial banks by means of three types of credit scoring models, namely, logit model, Probit model and linear discriminant analysis. The authors have derived the three models for estimation of the probability of default from a sample of 298 American commercial banks. Later on, authors tested the statistical significance of estimated parameters. After these models, the control sample of 100 American commercial banks is selected to determine the most suitable models. It is important to notice the evaluation of the estimated model and therefore their use and limitations. Three prediction models are estimated from the data set received during the financial crisis and this is limited to their use at this particular stage of evolution of the market. From this perspective, the estimated models are one-period prediction models with a rather short time prediction of default (1–2 years). The authors have further estimated that logit model is the most appropriate model for the prediction of banks’ default even if it has some limitations.
The authors have taken a sample of 100 American commercial banks, which were used to determine the most suitable model presented by Mileris (2010). It is important to calculate the estimated models and therefore their use and restrictions. Three forecast models are estimated from the data set received during the financial crisis and these markets are limited to their use at this particular stage of evolution. A total of 84 per cent of companies are properly classified by discriminatory activities. Customers classified by these functions as reliable must be rejected as not reliable credit applicants if they have ratings B, C and D. The article is conditioned by the high posterior probability of default.
In order to predict the bank risks research by Popescu (2014), the company uses the Altman’s Z-score model to discuss the use of discrimination analysis to assess the risk of bankruptcy. In this study, a case study was conducted in Romania, and three representative organizations related to dairy farming are chosen. Their financial statements, proportions and mainly Z-score are used for discrimination according to the standard. Results have shown that the profit of the farming companies related to dairy farms is low and bankruptcy risk levels are high. For this reason, the managers should keep monetary indicators under control at every moment and take necessary steps to recover at the end of the period so that failure will not happen to the company. Moreover, Kočišováa and Mišankováa (2014) has highlighted the importance of prediction models which are able to assess the financial health of the company which will give a warning signal to the stakeholders of the company. Authors use advance econometric methods like discriminant analsis, which is divided into univariate and multiple discriminant analysis (MDA). Authors also gave relevant importance for interpretation of differences, that is, canonical discriminant analysis and it presents statistical significant results.
Khatri (2016) has developed a unique discriminant function and it helps in generating warning signal about expected bankruptcy in advance of the actual bankruptcy. The score has given most importance to current ratio, asset turnover ratio and proprietary ratio. Study concluded that discriminant function is used with financial ratios as independent variable, instead of individual financial ratios, which can help in proper diagnosis and analysis of financial performance. Application of discriminant function for calculating Z-score can help in creating cautionary indicator about probable bankruptcy.
The study conducted by Mazilescu (2017) emphasized that the conditions related to the presence of certain similar financial features were only mentioned for applicability in the evaluation of the bankruptcy of the selected field and the use of certain initiatives in the field of activity. Here, the discriminatory analysis presents important differences between the two group initiatives (bankruptcy and non-bankruptcy) for each proportion of the recruitment. This analysis helps to understand if a sample is used to establish a financial performance assessment model. The study assists to choose between two methods to select variables of models: The addition of all the ratios that are allowed for further classification of two group categories and further selection based on statistical criteria or electoral inclusion. Possible ratio, on a priority basis, for example notoriety in literature, gives a subject-based character to select model variables. The individualistic character of the model is given by the motto of the model, such as the performance of models or the separation of bankruptcy companies. The separation is the preferred method of financing which uses asset structure, own resources and borrowed resources, and short-, medium- and long-term resources as the leading cause of bankruptcy. The results are confirmed by comparative analysis but are not complete as because of the risk assessment of the same company. The proposed models could be used only in the economies of those countries where statistical research was done, or in the activity sector, but the study was unable to expand their use across a larger area.
Madhushani and Kawshala (2018) have analysed the impact of financial distress on the financial performance of 29 listed ‘non-banking financial corporations’ of Sri Lanka. The study has identified that financial distress is having a significant impact on the financial performance of ‘financial institutions’. Altman’s Z-score has come to be positively related to return on assets (ROA) and return on equity (ROE) while leverage has negatively related to ROA and positively related to ROE.
Whereas, the article by Demyanyk and Hasan (2010) summarizes the study of empirical economics and operation research aimed at explaining, predicting and removing the financial crisis and bank failure in the United States and other countries. The article provides an analysis of the financial situation associated with the crisis in the subprime mortgage crisis in the United States. The study provides a comprehensive review of the intelligence techniques used in the research literature that facilitate in forecasting the failure of banks. However, Bhunia (2010), in his study, highlighted the degree to predict company failure using MDA by using 16 financial ratios on 64 Indian companies. The model has a tremendous predictive ability with above 80 per cent accuracy. The liquidity and profitability ratio has the highest success as predictive ability. Moreover, Platt and Platt (2006) have explained the difference between financial distress and bankruptcy. The study has highlighted the statistical difference between financial distress and bankruptcy prediction calculation; it shows that financial distress is more difficult to calculate than the bankruptcy prediction. The financial distress gives emphasis on operating decision, and debt levels are a guiding factor for bankruptcy. With the emphasis on India, the study by Pal and Bhattacharya (2013) has considered 15 financial ratios such as return on investment (ROI), debtors turnover ratio (DTR) and fixed asset turnover ratio (FATR). The study has judged the financial distress of Indian steel companies through these ratios. The study statistically signifies the importance of fixed asset management and good debtor management system so that profitability and ROI could be maintained at a high level. Therefore, the profitability and operation efficiency judging ratios have come out as the deciding criteria in predicting the financial health. Moreover, Stenback (2013) has provided an entirely new dimension to the traditional literatures of discriminant analysis. The study emphasizes that apart from the role of financial ratios in the discriminant analysis, some microeconomic variables are giving a new dimension to the predictive models. The microeconomic variables such as investment and export are included in the present study.
Therefore, by literature review, the authors understand that the Z-score model by Altman (1968, 2002) is most applicable and is widely utilized by the researches across the countries for analysing the financial stress position of the companies. Still, only this model is not sufficient to conclude the result and position of any organization. According to that, the authors move further into the research methodology section of this article.
Research Methodology
To realize our research objective that is analysing the financial performance of oil drilling and exploration sector of India and its affection by financial distress, we decided to utilize Altman’s Z-score analysis process and to validate the result of Altman’s process, discriminate technique will applied.
Research Design
We have compiled the financial data of 12 companies over the period of 2013–2017. The data used in our study are obtained from prowess database. The variables we have applied are working capital/total assets, retained earnings/total assets, earnings before interest and taxes/total assets, market value equity/book value of total debt, sales/total assets. For a description of the variables, see Table 1.
Description of Variables and Symbols
Z-score Analysis
We have utilized the Altman’s Z-score as our estimation instrument to assess an organization’s financial status. This linear compilation is compared to its pairs, an indicator of a given company that is known as Z-score; illustrate a better measure of risk or performance status.
Z-score can be formally formatted as follows:
Where, a0 is a constant quantifying the unexplained part of the score; R1, R2, … Rn are financial indicators (e.g. profitability, capital structure, risk and so on) of the selected companies and are assumed to be independent among themselves; a1, a2, … a represent the elasticity coefficients of the Z-score to unit changes in financial indicators and are estimated through simple least squares estimation.
The Z-score aims to develop a reliable scoring function which can very well discriminate companies into ‘healthy’ or ‘unhealthy’. The analysis depends on the financial parameter such as profitability, liquidity and risk which will calculate scores that are independent of these two groups. In particular, we are looking for a scoring function that increases the average scores of the good company and the average scores of the bad company and, at the same time, the quality of each group’s score reduces the deviations. The Altman (1968) has utilized five financial ratios to calculate the Z-score. These five ratios are already discussed in Table 1. We have utilized the linear function equation of Z-score from the same literature, which is as follows:
Where all the symbols have their above-mentioned variable indicator. Moreover, the Z-score and the variables X1–X4 are computed as absolute percentage values while X5 is computed in a number of times. The reader is cautioned to utilize the model in the appropriate manner. The variables X1–X4 are calculated on the absolute percentage terms. For example, any company has its net working capital to total assets (X1) is 10 per cent then it should be calculated as 10 per cent and not as 0.10. Moreover, the X5 (that is sales to total assets) is expressed as a ratio of 200 per cent that should be included as 2 (Altman, 2000).
The cut-off point of Z-score, which has come out from the above model, for a healthy companies ranges above 2.675 and for unhealthy companies, this score ranges below 2.675 (Altman, 2000). From Table 2, it is clear that 7 out of 12 companies are having their Z-score above the cut-off value of Z-score. The companies such as Deep Industries Limited, Petronet LNG Limited, Indian Oil, GAIL (India) Limited, Alphageo (India) Limited, ONGC and Dolphin Offshore Enterprises (India) Limited are having good health as far as financial risk is concerned whereas the companies such as SELAN, GSPC, Aban offshore, HOEC and Cairn India are having some financial risk.
Analysis and Result of Z-score
Discriminant Analysis
Discriminant analysis has calculated a set of equations based on independent variables that are utilized to discriminate the individuals into clusters. There are two objectives of discriminant analysis, the first one to compile a predictive equation for clustering the new individuals and the second one is interpreting that equation for predicting the relationships among the variables. Mathematically, discriminant analysis is interrelated with the one-way multivarate analysis of variance (MANOVA). In fact, variables have a reverse role in the method. The factors in the MANOVA are dependent variables in discriminant analysis and vice versa (Field, 2005). The discriminant analysis has explained the linear function of variables with the power of determined groups in accordance with selected characteristics (Laitinen & Kankaanpää, 1999). The discriminant analysis designs as a discriminant function that maximizes differences between centroids around selected variables (Pinches & Trieschmann, 1977). The quality of the designed discriminant function is shown in the accuracy of classification in one of the modalities of the categorical variable.
The discriminant analysis is conducted on the selected sample companies to predict group membership and to forecast the overall industry condition by distinguishing the sample companies in two groups—financially strong (healthy) and financially weak (unhealthy). This technique is used to classify companies into one of the alternative groups based on the set of interpreter variables. In this study, two groups (financially strong/financially weak) are to be compared based on the ratios mentioned above. The stepwise technique is retained to find out the combination of variables with the purpose of presenting the highest prediction accuracy rate and two variables came out to create the discriminant function.
Pooled within-groups matrices, —Table 3 statistically proves that X1 has the highest degree of positive correlation with X2 about 55 per cent and has a negative correlation with X4 about 49 per cent. X2 has the highest degree of negative correlation with X4 about 97 per cent. X3 and X5 have the lowest degree of correlation with other variables.
The functions potency and the abilities to discriminate are assessed by observing the eigenvalues, Wilks’ lambda, and chi-square with its significance level. Table 4 clearly shows that our discriminant functions explained the substantial portion of the variance as the function explained high eigenvalue, that is, 1.03.503 with Wilk’s lambda value 0.010 (Field, 2005). Moreover, the canonical correlation of the discriminating function is 0.995 that is significant at the 1 per cent level.
Correlation of All Ratios
Summary of Canonical Discriminant Functions
2. *Denotes 1 per cent level of significance.
It is clearly observed that 99.00 per cent, that is, square of the equations canonical correlation of the variations in the model between healthy and unhealthy function will change due to the five financial ratios of the companies. From Table 5, it is also understandable that the X1 ratio greatly contributes to discriminate the companies as financially healthy or unhealthy.
Standardized and Unstandardized Canonical Discriminant Function Coefficients
From Table 6, the group centroids of both the groups that are healthy and unhealthy are obtained. Risk factor line can be summed up as follows:
Functions at Group Centroids
Now in Table 7, it is clearly observed that the discriminant functions are successfully classified 75 per cent of original cases. In the present study, the standardized coefficient for classification is maintained by the average of the group centroid and the companies having a discriminant score less than average of group centroid are classified as financially unhealthy companies, whereas others are classified as healthy by the financial point of view (Field, 2005).
Classification Results
Results and Discussion
In the present study, the authors have estimated the financial distress of the 12 selected oil and gas companies of India. For that, the authors have first utilized the Altman’s (1968, 2000) and, Altman, Sabato, and Wilson (2008) Z-score model where the estimation shows that 5 out of 12 companies have financial risk as the model has explained the 5 well-known financial ratios (see Tables 1 and 2). The cut-off value for measuring the discrimination between the two groups of companies is clearly mentioned by Altman (2000), that is, 2.675. Moreover, only by one simple process of calculation, we could not conclude with our result and for revalidate our result for any conclusion, further discriminant analysis process were applied.
In the discriminant analysis, we observed that the Wilk’s lambda is very low and eigenvalue is very high which means the process of function estimation by the discriminant analysis is correct (see Table 4). Next, by calculating the unstandardized canonical discriminant function coefficients, we again build a model for financial distress of the selected sample companies and here we have observed that 9 out of 12 companies are correctly correlated with the previous result of Altman’s Z-score analysis (see Table 8).
Discriminant Scores of Sample Companies
Conclusion
Financial indicators were examined by discriminant analysis for the classification of ‘healthy’ and ‘unhealthy’ companies. The chances of financial distress for 12 companies using Altman’s Z-score over a period of 5 years from 2013 to 2017 have been analysed. The study reveals that 75 per cent of the selected companies are in financially healthy zone after the calculation of Z-score. The discriminant functions strength and ability to discriminate the selected companies is statistically assessed through the eigenvalues, Wilk’s lambda and chi-square. The X1 ratio, that is, working capital/total assets can very well discriminant the companies as financially healthy or unhealthy. In the discriminant analysis, we observed that the Wilk’s lambda is very low and Eigenvalue is very high that means the process of function estimation by the discriminant analysis is correct. Further, by calculating the unstandardized canonical discriminant function coefficients, we again build a model for financial distress of the selected sample companies and here we have observed that 9 out of 12 companies are correctly correlated with the previous result of Altman’s Z-score analysis. The past research reveals that in India, the study on Z-score is very few, so present study holds huge importance.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.