I had the wonderful opportunity presented in college to attend a 3 day workshop of Business Intelligence analysis and reporting using SPSS software. The reason I say wonderful is because research is a very big field and contributes a lot to the success of many big and small organizations. A lot of time and effort is spent by the researchers in back offices supplying valuable information to the front end and I was about to learn this piece of software called SPPS which would actually unweil to me the amount of detailing and also how things can be made simplified and beneficial to the organizations aspect clear to me.
On the first day I came 2 know that there basically are 2 views – the variable view and the data view – both having their own uses and offcourse the output really simplifies decision making. We were explained all the various tabs, headings, rows, labels in the views. We had hands on experience on working on the files that have been stored in the spss folder that we installed. We used them to analyze. We first worked on the cross tabs and frequencies feature. I really enjoyed working on the software.
Day 2, it was on perceptual mapping on a software called permap and also cluster analysis
Day 3, on discriminant analysis and factor analysis.
The best part was that like Im doing now, blogging on it. I thought instead of the 3rd blog being on some topic done on the 3rd day I end up blogging on how fun it is blogging about spss. It was really challenging workshop 2 , I mean required loads of concentration though at times I guess It was ok to while away a bit. The best part in doing the assignments we actually end up reading a lot of cases and seeing how the applications of SPSS and business analysis are being used in day to day lives. Few topics left and very much looking forward to it and also probably a career in research
And ya since its before the deadline of the submission of all the assignments, I really had fun blogging so far..
5.2.11
My experiences in the BI workshop – K.SAILESH, Roll no 12026, Marketing
I had the wonderful opportunity presented in college to attend a 3 day workshop of Business Intelligence analysis and reporting using SPSS software. The reason I say wonderful is because research is a very big field and contributes a lot to the success of many big and small organizations. A lot of time and effort is spent by the researchers in back offices supplying valuable information to the front end and I was about to learn this piece of software called SPPS which would actually unweil to me the amount of detailing and also how things can be made simplified and beneficial to the organizations aspect clear to me.
On the first day I came 2 know that there basically are 2 views – the variable view and the data view – both having their own uses and offcourse the output really simplifies decision making. We were explained all the various tabs, headings, rows, labels in the views. We had hands on experience on working on the files that have been stored in the spss folder that we installed. We used them to analyze. We first worked on the cross tabs and frequencies feature. I really enjoyed working on the software.
Day 2, it was on perceptual mapping on a software called permap and also cluster analysis
Day 3, on discriminant analysis and factor analysis.
The best part was that like Im doing now, blogging on it. I thought instead of the 3rd blog being on some topic done on the 3rd day I end up blogging on how fun it is blogging about spss. It was really challenging workshop 2 , I mean required loads of concentration though at times I guess It was ok to while away a bit. The best part in doing the assignments we actually end up reading a lot of cases and seeing how the applications of SPSS and business analysis are being used in day to day lives. Few topics left and very much looking forward to it and also probably a career in research
And ya since its before the deadline of the submission of all the assignments, I really had fun blogging so far..
On the first day I came 2 know that there basically are 2 views – the variable view and the data view – both having their own uses and offcourse the output really simplifies decision making. We were explained all the various tabs, headings, rows, labels in the views. We had hands on experience on working on the files that have been stored in the spss folder that we installed. We used them to analyze. We first worked on the cross tabs and frequencies feature. I really enjoyed working on the software.
Day 2, it was on perceptual mapping on a software called permap and also cluster analysis
Day 3, on discriminant analysis and factor analysis.
The best part was that like Im doing now, blogging on it. I thought instead of the 3rd blog being on some topic done on the 3rd day I end up blogging on how fun it is blogging about spss. It was really challenging workshop 2 , I mean required loads of concentration though at times I guess It was ok to while away a bit. The best part in doing the assignments we actually end up reading a lot of cases and seeing how the applications of SPSS and business analysis are being used in day to day lives. Few topics left and very much looking forward to it and also probably a career in research
And ya since its before the deadline of the submission of all the assignments, I really had fun blogging so far..
Risk Classification for SMEs : An application of Discriminant Analysis
Credit rating agencies specializes in analyzing and evaluating the creditworthiness of large corporate and issuers of debt securities. In the new financial architecture, credit rating agencies are expected to become more important in the management of both corporate and credit risk. Their role is limited to the large scale companies and multi corporations. Credit Rating Agencies focus was never on Small and Medium Enterprises where credit worthiness related information asymmetry is too large. On the other hand banks also handicapped by not having robust comprehensive models. To bridge the gap this research attempt has been made to provide solutions to the small and medium enterprises and banks.
I. INTRODUCTION
In India, primarily the concept of Small Scale Industry has been in vogue and the medium enterprise definition is of more recent origin. An SSI is defined on the basis of limit of historical value of investment in plant & machinery, which at present is up to Rs.10 million. However, in respect of some specified items, this investment limit has been hiked to Rs.50 million. For the recently announced Small and Medium Enterprises Fund, the Government of India has approved the limit of investment in plant and machinery above Rs.1 crore and up to Rs.10 crore for defining a unit as a Medium Enterprise. Amongst the developing countries, India has been the first to display special consideration to SSIs and basic focus has been to make economical use of capital and absorb the abundant labour supply in the country.
India it is often stated that 60 percent of SMEs do not borrow from traditional sources. For those who borrow from traditional sources the question arises of what measures should be used to assess applications for loans. Most of Small and Medium Enterprise operate in very small scale with very limited equity of the owner and more on high cost debt fund from other sources such as external borrowing from Non Banking firms.
Small and Medium Enterprises (SMEs) constitute a significant part of developing economies, this was emphasized in the research works of Zolton ACS & Audretsch (1993), OECD SMEs Outlook (2002) and Allen N. Berger & Gregory F. Udell (2004). Majority of these enterprises fund their capital through family or other networks, a sizeable group will borrow from traditional suppliers of credit.
Taffler, (1982) “Forecasting Company Failure in the UK Using Discriminant Analysis and Financial Ratio Data In the modeling of default using Accounting based approach within this paper one has extended the range of variables considered and applied standard Credit Scoring approaches in modeling, see Lin, Ansell & Andreeva (2007).
Dr. V.Manickavasagam and Srinivas Gumparthi (2009) A Risk Assessment Model (RAM) is necessary to avoid the limitations associated with a simplistic and broad classification of applicants into a "good" or "bad" category. The absence of appropriate weights in the current evaluation system triggers the need for the development of the comprehensive model based on proven statistical application. Literature survey undertaken brought to surface 28 parameters that need to be taken into account while evaluating a prospect. These parameters were classified under four heads namely credit, operations, liquidity and market risks. Weights developed in this study were based on a conceptual understanding and the importance attached by people proficient in this area. A questionnaire was developed and a judgmental survey was conducted for this purpose amongst various credit officers extending commercial vehicle and construction equipment financing. The sample size was 117 small and medium corporate clients.The existing model was able to classify 28 records correctly. So the predictive power of the original/existing model was about 80%. The proposed/new model is able to classify 30 records correctly. So the predictive power of the propose/new model is 85.71%.
II. NEED FOR COMPREHENSIVE RISK ASSESSMENT MODEL
There is need for simple and easy to understand credit rating model for SMEs.’ At present SME’s are depending on either third party agency rating or at the mercy of the Credit Managers of the bank. To design and develop risk classification model based on Discriminanrt Analysis for Small and Medium Enterprises (SMEs’) which are major contributors to the growth of Indian Economy. Primary objective is to develop risk classification model based on Discriminant Analysis for Small and medium enterprise (SMEs’). Discriminant Analysis is a proven statistical application. Inference of the model is easy to understand and simple for implementation.
Limitations of the Study: The purview of the project is limited to Small and Medium Enterprises (SME) division. Only twenty parameters are used in developing this model. Parameters selected are indicative in nature and by using them we can clearly establish the intended results.
III. DISCRIMINANT ANALYSIS MODEL
Based on developed Credit Risk Frame work, Clients are classified in to various categories depending on the aggregate score. For further simplification of the under two risk categories such as good and bad assets the Discriminant Analysis has been applied. This is second stage of the total new model. By using model Credit Manager is able to see clear distinction between performing assets and non performing assets. Statistically application of Discriminant Analysis purpose is to classify objects/records into two or more groups based on the knowledge of some variables related to them.
Discriminant Function
Y = a+ k1X1 +k2X2 +………. + knXn
Where Y-Dependent variable, a – Constant, X1, X2…Xn- Independent variables, k1, k2- Coefficients of the independent variables.
In this case, for the development of the model the dependent and independent variables are as follows
• The dependent variable (Y) is the Client Risk Rating (CRR)
• The independent variables (X1, X2……X20) are as follows
X1 = Client history, X2 = Industry status, X3 = Relationship with suppliers, X4 = Relationship with customers, X5 = Competition, X6 = Liquidity, X7 = Leverage, X8 = Sales growth, X9 = PBDIT/sales, X10 = DSCR (Debt service coverage ratio), X11 = Integrity, X12 = Family standing, X13= Financial standing, X14 = Management competence, X15 = Management commitment, X16 = Succession, X17 = Employee quality, X18 = Internal controls, X19 = Repayment records, X20 = Compliance records
For designing and development of Discriminant Analysis statistical application; the value of the dependent (credit score of individual firm based on credit frame work) and the 20 independent variables for the 70 records are entered in the SPSS software. The dependent variable in this equation Client Risk Rating based on the credit score obtained from Risk assessment format of Risk Assessment Model for Assessing NBFCs’ (Asset Financing) Customers of Dr. V.Manickavasagam and Srinivas Gumparthi[8] (Stage – 1 of the model). The Client Risk Rating was computed on the basis of successive seven years data of the client from 2002-03 to 2008-09. The period has importance in this model because of it cyclical nature. The Economic cycle has influenced business cycle of small and medium enterprises in this mentioned period. The Gross domestic production rate fluctuated from 2002-03 to 2008-09. Initially GDP pegged up from 2002 -03 to 2006 -07 and declined after that due to global economic slow down impact. During mentioned period, small and medium enterprise across the all industry categories followed the same trend. Except Auto and Auto Ancillary sector all other categories taken for studies were impacted by the economic cycle during the period. During the mentioned period it was observed that GDP rate from 2004-05 to 2006-07 was at peak in India over 8 to 9%. Due Economic Slow down even Indian has seen decline in GDP rate below 6% during 2007-08 and 2008-09. The clients who are selected for building this model were with the existing bank through out this period survived at least 24 quarters. Data compiled on continuous basis and obtained on all parameters mentioned in the Risk Assessment Framework. All the observations and out come of the credit score were checked for its accuracy and consistency before formulating into Discriminant Equations. The Discriminant Scores are computed by solving all the 70 equations.
TABLE. NO: 1.ANALYSIS CASE PROCESSING SUMMARY
From the above results it is clearly evident that out of 70 SME clients, 64 are performing clients and 6 are defaulters.
IV. DATA ANALYSIS BASED ON DISCRIMINANT ANALYSIS
The Second phase of this model application was using Discriminant Analysis for further validation reinforcement of model robustness. By using this analysis all 140 clients were classified as Performing Assets and Non Performing Assets. By using range sub classification method, the Performing Assets are further regrouped under High Risk, Medium Risk and Low Risk, which is useful for monitoring progress of the Client.
Among the performing assets approximately 20 percent of SME clients are in high risk category, which is concerned issue to the bank. These four clients are to be closely monitored. Further diagnosis is possible through this model to identify exactly what type of sub classification risk each of this clients are facing. There are 8 clients in the category of medium risk most of them in this position due Economic slow down which was identified through their market risk analysis (Economic Cycles) of the proposed model. Since all clients are from SME sector such type analysis helps them in knowing their risk level.
Above table states that 6 out of 25 Performing SME clients are in High Risk Category and 10 Assets are in the Medium Risk Category. On further probing the High Risk Category assets it was observed that these clients very small to with stand the competition especially when economy has slow down considerably in 2007-08 to 2008-09. But there is fair chance doing well when turn round happens in the business.
SME clients belongs to Engineering in the high risk category are facing Business risk due to slow down in economy. No fresh order in from the market has further complicated their business model. In depth analysis shows that 2 assets of Textile and Apparel industry category are facing high risk, which is mainly due slump in exports to European Countries. Past Record of the company was very good export performance.
SME clients of engineering industry category were facing high risk and two other are in Non Performing assets class. This clearly shows that SMEs’ are facing severe impact of economic slow down. These companies have to be internally strong enough to sustain during this face till they get new orders. This analysis is possible due the accurate risk class of the client.
From analysis it is clear that 50 percent of bank’s SME clients are either in Non Performing assets category or in high risk class. This classification is useful for the bank to formulate it strategies to over come default issues. Bank has to reformulate its lending portfolio composition. Further diagnosis will help in guiding their SME clients.
All 140 samples considered for this study has been first categorized into Performing Assets and Non Performing Assets based on their performance by using discriminant analysis. Further sub classification was done. The model capability was demonstrated through this classification of assets, which is third added feature of this model, not many models which are available in the market have three important features such rating, discrimination between good and bad assets and risk classification.
Submitted by;
Sumit Acharya
Roll - 12108
Finance Batch
SIBM Bangalore
Factor Analysis-Client Financial Institution
Factor Analysis attempts to identify underlying variables (factors) that explain the pattern of correlations within a set of observed variables. Factor analysis is often used in data reduction. It can also be used to group variables (e.g. Bank Transaction Preferences) into discernable patterns.In this research paper two Factor Analyses were performed. The first on Question XX to attempt to group preferable methods of bank transactions together. The second factor analysis was performed on Question of financial reasons into XX, which groups the important clusters that the bank may want to use to identify customer needs and create new services to address them.The output is sorted in descending order of factor loading. Variables are reported with the loading of its factor rounded to two decimal places. All loadings below 0 .4 are omitted.
Banking Transaction Preferences Group into Three Distinct Methodologies:
Customer’s Primary Banking Concerns Can Be Broken Down into Five Main Categories:
Refer Table below.
Refer Table below.
Hence to sum it up Customer Preference for Banking Transactions Can Be Broken Down Into Three Methodologies; Computer/Phone, Remote, and Face-to-Face. This is useful information for the bank because they may now be able to group customers according to their preferred transaction method and offer services accordingly.Customer’s Primary Banking Concerns Can Be Broken Down into Five Main Categories. Again, Some Banks have an opportunity to understand that various banks can be grouped together. For example, if a customer indicates that he is using the bank to that he may be interested in paying off his mortgage more quickly. This is known because these reasons fall in the same factor and have a high correlation. Customers can also be clustered in this manner.
Submitted by;
Dibyangana Saha
Roll - 12132
Finance Batch
SIBM Bangalore
Strengths and weaknesses of Discriminant Analysis
The simplest (and usually very poor) parametric classification technique consists in assigning an observation to the class whose barycenter is closest to the observation, in the Euclidian sense of the term. This is a "first order" technique, in that in considers only the barycenters of the classes, whose coordinates are linear (first degree) functions of the observations, and does not take their spreads in the various directions of space into consideration.
The next step is to take the "second order information" about classes into account, that is, to consider their covariance matrices in addition to their barycenters : the coefficients of a covariance matrix are quadratic (second degree) functions or the observations . This is exactly what Discriminant Analysis does. In fact, rather than emphasizing the "class normality" assumption, it is equally justified to say that Discriminant Analysis is the simplest "second order classification technique", with then no reference to a particular functional form of the class densities. But because a multinormal density is fully determined by its mean and its covariance matrix, Discriminant Analysis is the ideal classification technique for multinormal classes.
The beauty of it is that it will turn out that Discriminant Analysis may still be interpreted as a "nearest barycenter" technique (first order), provided that the usual Euclidian distance be replaced by another "distance", the Mahalanobis distance, which is defined from second order information about the classes.
So the strength of Discriminant Analysis is that it is an optimal yet simple, complete and accurate classification technique whenever the basic "second order" assumptions accurately describe the setting (multinormal classes).
Of course, this is also where its weakness is. Classes are never perfectly multinormal : to what extent do the performances of Discriminant Analysis degrade when the class densities depart form normality ? Theory has no definite answer to this question, but one of the reasons for the popularity of Discriminant Analysis is that decades of intensive use show that DA is reasonably robust with respect to departure from the standard assumptions.
Another weakness of the full Discriminant Analysis is that as many covariance matrices have to be estimated as there are classes. This easily leads to models containing tens or even hundreds of parameters, a rather large number in view of the usually limited amount of design data. Consequently, the full DA tends to be unstable (strong dependence of design data), whereas the "restricted" versions, with fewer parameters, tend, to the contrary, to be more stable but at the expense of a strong bias. Modern versions of Discriminant Analysis incorporate regularization mechanisms, whose principles are similar to that of Ridge Regression, and which allow fine tuning of the balance between bias and variance of a DA model.
Ankit Agarwal, 12070, Marketing
Ankit Agarwal, 12070, Marketing
Confirmatory Factor Analysis
Confirmatory factor analysis (CFA) is a multivariate statistical procedure that is used to test how well the measured variables represent the number of constructs. Confirmatory factor analysis (CFA) and exploratory factor analysis (EFA) are similar techniques, but in exploratory factor analysis (EFA), data is simply explored and provides information about the numbers of factors required to represent the data. In exploratory factor analysis, all measured variables are related to every latent variable. But in confirmatory factor analysis (CFA), researchers can specify the number of factors required in the data and which measured variable is related to which latent variable. Confirmatory factor analysis (CFA) is a tool that is used to confirm or reject the measurement theory.
Terms and concepts in confirmatory factor analysis (CFA):
· Theory: In confirmatory factor analysis (CFA), theory is a systematic set of causal relationships that provide the comprehensive explanation of a phenomenon.
· Model: In confirmatory factor analysis (CFA), model is a specified set of dependant relationships that can be used to test the theory.
· Path analysis: In confirmatory factor analysis (CFA), path analysis is used to test structural equations.
· Path diagram: In confirmatory factor analysis (CFA), the path diagram shows the graphical representation of cause and effect relationships of the theory.
· Endogenous variable: In confirmatory factor analysis (CFA), endogenous variables are the resulting variables that are a causal relationship.
· Exogenous variable: In confirmatory factor analysis (CFA), exogenous variables are the predictor variables.
· Confirmatory analysis: In confirmatory factor analysis (CFA), confirmatory analysis is used to test the pre-specified relationship.
· Cronbach’s alpha: In confirmatory factor analysis (CFA), Cronbach’s alpha is used to measure the reliability of two or more construct indicators.
· Identification: In confirmatory factor analysis (CFA), identification is used to test whether or not there are a sufficient number of equations to solve the unknown coefficient. In confirmatory factor analysis (CFA) identifications are of three types: (1) underidentified, (2) exact identified, and (3) over-identified.
· Goodness of fit: In confirmatory factor analysis (CFA), goodness of fit is the degree to which the observed input matrix is predicted by the estimated model.
The following are the procedures involved in confirmatory factor analysis (CFA):
1. Defining individual construct: In confirmatory factor analysis (CFA), first we have to define the individual constructs. In confirmatory factor analysis (CFA), the first step involves the procedure that defines constructs theoretically. This involves a pretest to evaluate the construct items, and a confirmatory test of the measurement model that is conducted using confirmatory factor analysis (CFA), etc.
2. Developing the overall measurement model theory: In confirmatory factor analysis (CFA), we should consider the concept of unidimensionality between construct error variance and within construct error variance. At least four constructs and three items per constructs should be present in the research.
3. Designing a study to produce the empirical results: In confirmatory factor analysis (CFA), the measurement model must be specified. In confirmatory factor analysis (CFA), most commonly, the value of one loading estimate should be one per construct. In confirmatory factor analysis (CFA), two methods are available for identification. The first is rank condition, and the second is order condition.
4. Assessing the measurement model validity: In confirmatory factor analysis (CFA), assessing the measurement model validity occurs when the theoretical measurement model is compared with the reality model to see how well the data fits. In confirmatory factor analysis (CFA), to check the measurement model validity, the number of the indicator helps us. For example, in confirmatory factor analysis (CFA), the factor loading latent variable should be greater than 0.7. Chi-square test and other goodness of fit statistics like RMR, GFI, NFI, RMSEA, SIC, BIC, etc., are some key indicators that help in measuring the model validity in confirmatory factor analysis (CFA).
Ganesh Singh, 12021, MBA Marketing
Subscribe to:
Posts (Atom)