Brand Perception of Facebook, Twitter and Linkedin via Perceptual Mapping

Perceptual Mapping is something I learned from Proff. Uday Bhate in my Business Intelligence class and it is pretty interesting. It is a way to get consumer insight from the consumer by having them actually draw on a map. Perceptual mapping helps to communicate the relationship between competitors and the criteria used by consumers while making purchase decisions. Perceptual maps, being simple graphic figures, can pave the path towards strategic thinking and trigger discussions. Rating scales provide the data required for perceptual mapping. These rating scales will have the subjects of the map described on the basis of selected attributes. The term of validity of the map depends on the overall set of attributes as well as the subjects.

The Map has the following Characteristics:

1) Pair-wise distances between product alternatives directly indicate how close or far apart the products are in the minds of customers.

2) A vector on the map indicates both magnitude and direction in the Euclidean space. Vectors are usually used to geometrically denote attributes of the perceptual maps.

3) The axes of the map are a special set of vectors suggesting the underlying dimensions that best characterize how customers differentiate between alternatives.

Objective: of this blog post is to study in very simple terms how the three stalwarts of the social media space – Facebook, Twitter and Linkedin are perceived by a common user (SIBM Students).

Purpose: of carrying out this analysis is to debate whether there is a gap in the respective company’s desired positioning vis-à-vis the actual perception.

Method: Perceptual Maps are easy to interpret graphs that visually display the perceptions of customers about a brand. I have used this technique to display the relative position of Facebook, Twitter and Linkedin across certain key attributes.

Analysis by way of Perceptual Maps:

Attribute 1: Social Networking
Attribute 2: Professional Networking

Results:

• Facebook scores average on personal networking as well as professional networking.
• Facebook users tend to look for and interact more with old friends/acquaintances.
• As less popularity among users twitter score less on social as well as professional networking.
• Twitter users are more open to making new friends – who had been total strangers till they met on Twitter.
• Linkedin’s perception is that of a business/professional networking site. That’s how it has been officially positioned as well.
• Linkedin users also prefer to use this mode for social networking too.

Conclusion:

In the end with the help of perceptual mapping we can draw the meaningful conclusion about the perceptions of the people means what the people think about a particular brand. And with the help of the same companies can target the particular customers where they feel that they are lagging in a particular segment. According to my view, facebook is more about friends and family whom we know reasonably well and would have met some time in life. LinkedIn is basically about professional networking, primarily about ex and present colleagues. It’s a boon for corporate recruiters. Twitter is all about conversations, opinions, creativity, humour and brevity, content rules and it is a boon for celebrities.

So, on the basis of above findings we can say the perceptual mapping is the very powerful tool for analysing people’s perception in today’s era.

Submitted By:

Amit Kumar
12011
SIBM-Bangalore

Business Intelligence : Your automatic 'Strategic Choice' to stay ahead of the rest.

A course on the days ‘in-thing’ Business Intelligence and its applications right before joining the IT industry towards the end of one’s MBA curriculum does bear the promise to add a lot of value to the individual’s career profile and hence creates an umpteen environment of excitement.

So what is Business Intelligence all about? Internet Definition goes like this : Business intelligence (BI) refers to computer-based techniques used in spotting, digging-out, and analyzing business data, such as sales revenue by products and/or departments, or by associated costs and incomes.

In ordinary terms – BI technologies provide historical, current, and predictive views of business operations. It is related to Gathering information to solve a particular business requirement, data mining, generating reports and presenting it to the management to make necessary analysis and action decisions.

Thus an introduction to BI helped us grasp its basics, more so with hands on experience on SPSS and Permap software. SPSS is a computer program for statistical analysis, originally designed for Social Sciences. And Permap is a Multi Dimensional Scaling tool which helps in comparing between different attributes among samples based on their overall / attribute similarity.

The SPSS hands on makes one gain knowledge on;

A. 1^st Level Analysis (Arriving at business decisions based on frequency, Bi variate data analysis using Cross Tab functionality focussing on customers behavioural or personality traits using frequencies, percentages, Null Hypothesis testing & Chi Square Tests)

B. 2^nd Level analysis (Arriving at business decisions based on Cluster analysis – grouping the data based on clustering objective,use of Hierarchical or K-Means cluster depending on the no of cases involved, identification of Proximity Matrix and visual depiction using Dendograms).

The Permap hands on helped us get familiarized with the concepts of;

A. Multi Dimensional Scaling (Arriving at business decisions on minimizing perceptual error about objects like, brands or products etc using Perceptual Mapping to identify attribute based similarity and sensitivity analysis on the change of these decisions by altering one or more of these attributes).

B. Discriminant Analysis (Arriving at business decisions about certain aspects based on past historical data, predicting the characteristics of certain attributes, verifying the change in nature of such attributes based on conditionality and finding the accuracy of behaviour or decision predictions.)

With so much technicality involved what runs among most of us is the fundamental question, if BI is such a powerful tool where in lies its mass applications? Being such a hyped and well researched field of study what are some of the next big trends in this sector? What all are the tools & techniques are used in most organisations? To satiate my inquisitive itch I came out with 2 interesting industry applications wherein the usage of BI has probably transformed the game altogether,

1. Logistics - Over the last few decades the role of logistics management has undergone a paradigm shift & has become an extremely important aspect of the overall business strategy. The increased complexity of logistics management has led many companies to outsource their logistics activities to 3^rd Party Logistics (3PL) providers. Today, 3PLs play a critical role in the supply chains of their customers. They are increasingly viewed as strategic partners who can play a pivotal role in optimizing the supply chain and thereby providing sustained competitive advantage.

To effectively manage the supply chains of their customers, 3PLs need to constantly analyze data collected from various sources and convert it into actionable information. Business Intelligence tools like data warehousing and OLAP can significantly help 3PLs in achieving this objective. By providing a unified view of the entire supply chain, these tools can help improve the functioning of basic 3PL services like transportation management, warehousing and inventory management. 3PLs can leverage BI tools to provide their clients with information specific to their supply chain, thereby increasing their market responsiveness. BI tools also help 3PLs improve their own internal organizational functions like human resources and financial management.

(More info on this available in the White Paper : Business Intelligence & Logistics by Authors: Srinivasa Rao & Saurabh Swarup – BI/DW Consultants – WIPRO Technologies)

2. Banking - The banking industry has invested heavily in information technology and generally banks are quite sophisticated in their use of IT. However, when it comes to standardization and industrialization of systems and processes inside a bank, well, the industry lags behind its counterparts. Industry standards have already defined the foundation for business process interoperability in the banking industry, particularly with respect to the exchange of funds with other banks or financial networks. Towards this objective certain applications of BI prove to be of extreme value addition like the usage of XBRL - Extensible Business Reporting Language.

XBRL is a language for the electronic communication of business and financial data which is changing banking business reporting around the world. Companies use XBRL to save costs and streamline their processes for collecting and reporting financial information. Consumers of financial data, including investors, analysts, financial institutions and regulators, can receive, find, compare and analyze data much more rapidly and efficiently if it is in XBRL format.

Moving ahead I tried to gather an insight on the latest & upcoming trends in the BI practices arena. And here’s what the experts have to say on this;

1. One of the top trends for business intelligence in 2011 is the movement towards utilization of Social Data. There are over 250 major social media and networking sites producing data that can be incorporated into enterprise business intelligence environments.

2. Small Business. Business intelligence is getting simpler and easier to setup as more and more vendors try to enter the market. This will enable smaller companies to take advantage of business intelligence.

3. BI on mobile devices - getting closer to the 'right data, right information, right insight, right time, right person' mantra of BI.

4. End-User Driven Visualization. Large reports, even end-user driven ones are pretty useless if they just present large amounts of data in text format or similar. Customised visualisation & evolution of Intelligent intuitive tools that can understand the context of the data being presented and can choose one or a few of the most appropriate ways of displaying the information would emerge to be the call of the day.

What they really hope for is some significant forward movement in the BI arena - while each vendor slowly bringing us a little further forward, it does seem to be a very slow tide moving. Honestly we remain stunned when we reflect as to where we were on the Microsoft 2000 version or the Cognos 95 version and where these two major vendors are today - and compare their usage in today’s I-phone technology. Riding high on that dream voyage perhaps we may even dare to believe that 2011 maybe the year for bringing 'BI from business corporate to the masses'.

Submitted by;

Sumit Acharya

Roll - 12108

Finance Batch

SIBM Bangalore

Perceptual Mapping--A Boon For HR!!

It was the second day of our Business Intelligence (BI) Workshop and I was looking forward to it after the insights that we delved into after day 1 session on Clustering and Dendograms. The session was centered on the concept of Perceptual mapping and its usage in various functions of Business.
My biggest apprehension before taking up BI was whether it will add any value to me as I am from the Human Resource Specialization. The doubts were made crystal as soon as the sessions began. Starting from Clustering to Perceptual mapping, the usage of these are aplenty and in every domain of business.This was made very evident by our discussion and the examples discussed in class.

The topic that caught my attention today was that of Perceptual Mapping. The usage of Perceptual mapping was explained to us through a Mobile phone example where in we could map the perceptions that people have about the different Smartphone’s that are selling in the market. Hence, the perspective provided to us was from the Marketing angle. Being a student of HR, I tried to draw parallels between its usage in Human Resourcing and more importantly the areas that are a bane to the organization.

HR is currently in its best possible era wherein the firms are finally realizing the potential that lies in its effective utilization and the value it can add to the bottom line of the business. Thinking from the point of view of a budding HR professional, to add value to business, I think Perceptual Maps is a fantastic tool which can find application in solving/analyzing the various problems plaguing HR in firms today.

As an example, as per the PwC report, “Reconnecting HR to the Bottom line” ,today the biggest problem IT/ITes firms are facing is that of Attrition. The numbers are startling and CEO's around the world have rated this as their primary area of concern. Thus, an analysis on the trends prevalent in attrition is of paramount importance to any organization today.
As a methodology to finding out the reasons of employees quitting their jobs and drawing a correlation among them can enable us in stemming the exit of the employees and make the business more profitable.

The perceptual Map would allow me to analyze attrition from the perspective of different factors. I would be able to gauge the biggest contributor to why people are quitting their jobs and how I would be able to retain them. However this is not something new, companies have been doing this for years now.
The unique feature of the perceptual map is that it allows us to see the relationship that exists between the different factors also.
To understand this, let us assume that the 2 most important reasons for attrition are growth of the individual within the organization and the learning opportunities that the firm provides to the employee. So far we have never tried to find out if their exists any relation between these 2 stated factors, however through a perceptual map we can find out if these 2 are indeed related!!Hence, our perspective on the reasons for attrition increases manifold and curbing it would become relatively simpler.

As per a study on Attrition by Peter Capelli , a firm spends nearly 60% if its cost on employees, out of which, Training and Recruitment form the biggest chunk after Salary. Hence, attrition is a major contributor to the expenses of a firm. This is not all, high attrition numbers reduce the goodwill of the organization in the market.
Another aspect of Perceptual map that we can use is that of the comparison of attrition levels of the competitors in the market. This will give us a good idea of how the others are performing in terms of attrition. Lower attrition levels could be a source of competitive advantage to the firm and can drastically reduce the expenses that a firm undertakes in dealing with employees leaving the organization on a regular basis.

Hence, analysis of Attrition patterns in a firm can be understood well through Perceptual maps and I think I will make extensive use of it in various other divisions of Human Resourcing as well!!

Submitted By:
Ashish Bhatnagar
Roll no:12128 (HR)

Current issues and limitations faced while using multivariate analysis

After understanding the basics of Perceptual Mapping in class I thought it would be wiser for me to search evaluate and understand some of the current trends and applications of the above subject in the current world. Going deeper into this I encountered a lot of interesting facts shared below and which I think would help us in understanding this analysis in a better manner.Below are some of the details of the limitations of using multivariate analysis in perceptual mapping.

Current Limitations

There are a lot of limitations in the usage of the multivariate analysis in perceptual mapping.There three types of limitations that must be placed on the relevant multivariate space that would discuss are:

Limits on the relevant set of variables that will be used to define the perceptual space

This is one of the most critical area for setting limitations.(except for those using the scaling methods based on overall product similarities).The major question we need to answer is what variables are to be used to orient the perceptual positioning of the various competitors. There is a nearly unlimited set of variables available. The selection of the relevant variable set determines the type of map that will be produced. That is, will the map be based on such things as

• Purchase Behavior
• Organizational Images
• Product usage behaviors
• Product attribute characteristics
• Brand images
• Consumer goals & consumer needs
• Convenience issues
• Combination of the above or other parameters

Limits on the population that is to be surveyed

This seldom poses a serious problem because it tends to be self-defining in terms of users, or purchasers of the products, services, or firms in question. However, there are questions as to how familiar a respondent is with a product, or brand.

Limits on the relevant set of products, services, or firms

This is also a major issue. Although this is as critical an issue as the selection of the relevant variable set, it is still a serious one. A balance is required in this era of market fragmentation, and the rapid emergence of new product categories, and subcategories, brought on by an acceleration of differentiated products flooding the market place, the selection of the relevant competitive set of products or services is ever-changing.

Conclusion

There are still many issues and there is a huge further development opportunity with multivariate analysis we keep addressing these issues we believe this methodology will give us better and improved results.

Submitted By

Neetu Rathod

Roll No: - 12118

SIBM Bangalore

Ref:-http://www.sdr-consulting.com/article11.html

BI never lies... by Jins Jose (12085)

[Let me start with the disclaimer: Any similarity of character of the story to a real person is intentional]

Mr Nair is one of the youngest Relationship Managers in CITI bank. The bank has given him an extensive target of acquiring HNI (High Net worth Individual) accounts in the Maharashtra area. As you might know “CITI never sleeps”; so is Mr Nair too, while carrying extensive responsibilities.

Mr Nair is a representative of the Gen-Y Indian professionals who believe in the power of technology and market intelligence. He used to leverage Business Intelligence (BI) software while taking key business decisions.

Mr Nair started churning out the correlations between different economic and demographic factors associated with the population under his sales area. While evaluating different fields in the SPSS (Statistical Package for the Social Sciences) spreadsheet, he could find a surprising correlation between 2 fields- ‘occupation’ and ‘wealth’. Those with occupation as ‘farmer’ were showing a strong potential to be ‘prospective HNIs’. More surprisingly, these farmers owned less acreage.

So Mr Nair started with the null hypothesis: “there’s no chance that a farmer with acreage < 10 could be an HNI”.   He set his confidence level requirement to be 99%, i.e. he wanted ‘Pearson Chi-square’ value to be less than 0.01 to reject the null hypothesis showing the huge potential of those farmers to be HNIs.  For his surprise ‘Pearson Chi-square’ was even lesser than 0.01. His brain murmured Mr Nair: “how the hell it can happen?” while his heart was whispering: “BI never lies”.

It was almost impossible for Mr Nair to convince his boss and colleagues these new findings; but finally, somehow he could manage. Nair’s track record didn’t allow the boss to neglect him.

Months passed; it was the time for Mr Nair to deliver results rather than findings; he delivered. It was December 2010. Onion price was soaring to Rs.100/kg.  Mr Nair was given the best Relationship Manager award for the quarter for adding hundreds of onion farmers into the banks HNI network.

Perceptual Mapping

Today we had a lecture on perceptual mapping which helped us understand how the market researched data can be analyzed to get some logical conclusions. Below is a brief on the above topic.

Introduction

Perceptual mapping is a market research technique which helps us in mapping the perceptions of the objects in a diagrammatic manner which in turn helps us understand the relation between the positioning of researched object in market. It also helps us to take a broad view of the properties of the product compared with its peers. This helps us strategize as to what our next step should be.

The above technique has been in existence for over twenty years now. It is one of the few advanced multivariate techniques whose popularity has not been affected much over these years. This technique uses different algorithms to generate the outputs which are discussed below. Below are some of the algorithms that are generally in use for perceptual mapping

Discriminant analysis

It is one of the most popular algorithms in use today for applied multivariate mapping. The inputs to discriminant analysis consist of individual respondent ratings of products across attributes. The basic assumption here is that the rating scales are continuous and normally distributed. It is much like regression analysis in that it uses a least squares approaches in an attempt to fit linear models to the data. However, the dependent variable is nominal. That is, for mapping purposes, the dependent variable is the product being rated.

Thus, each product rated by each respondent is an input record, so if a respondent rated five products, which generate five input records. Discriminant analysis then calculates the coefficients to a set of standardized linear equations, called discriminant equations, which explain the differences between the product ratings. Or, said a different way, explains the variance between product ratings.

These algorithms provide a variety of useful statistics to the researcher, such as Eigen values to show you the variance explained by each equation, tests of significance for each equation, multivariate F statistics to show the significance of the group differences, and correlations between the discriminant functions and each attribute variable.

R-Type Factor Analysis

It is seldom used as a mapping procedure in today's MR Field. There are a few empirical studies though that shows it is superior to discriminant analysis. Although you have the same problems with what to do about missing data and selecting the relevant set of variables as we have with discriminant analysis, this procedure overcomes two of the problems with discriminant analysis. All variables are shown on the map, and the inclusion or exclusion of products has no effect on the extracted dimensions.

The inputs to factor analysis are very similar to those for discriminant analysis, product ratings across attributes. However, an additional ingredient is required; we must also collect an importance rating from each respondent for each attribute. These importance ratings are the basis for developing the mapping space. The basic assumptions concerning the distribution and continuity of the rating scales should not be relaxed.

Factor analysis is an interdependence procedure, thus the various differences in product ratings is ignored until after the factor equations are derived. Product locations in the derived space are calculated by averaging the first two factor scores of that product's ratings to define the X and Y coordinates. Or alternatively, plugging the average product scores on each attribute into the two factor scores and calculating the X and Y coordinates.

Ex: - We evaluated a mobile handset perception map which consisted of comparing five different models and understand there positioning.

Submitted By

Arun Ramakrishnan Roll No - 12127 SIBM Bangalore

Ref:-http://www.sdr-consulting.com/article11.html

Perceptual Mapping of food joints in and around e-city

To begin my blog on perceptual mapping and start my journey both in the space of blogging and SPSS I have decided to take a simple but useful exercise. The exercise shows how effectively we can use theory (Perceptual Mapping in this case) in our daily life and help us make better decisions.

To start with I decided to make Perceptual maps of food joints in and around our college which most of us frequently go and to find out what most of us perceived them on two parameters i.e. quality and price. I have limited the parameters to two in order to keep things simple but in future more parameters can be taken to analyze further the results we got from our initial responses. I have taken 5 food joints in and around our college and got response from students as what they rate them on Price and Quality to gauge what we called getting “Bang for our Buck” aspect of various joints. Survey was done on 15 students and parameters were rated on 1 (Low) and 5 (High) scale.

Attributes: 1- Price and 2- Quality

Results:

1) Price and quality even though should be on the same lines i.e. Quality should be consistent with the price, as quality should increases with price but the map shows that is not the case entirely.

2) Domino and Pizza-Hut is equi-distance from both the attributes i.e. Price and quality (both scores high on price and quality) but domino gives better quality even though it is equal in price as compared to pizza hut.

3) Punjabi Dhabba is last in quality even though its expensive than the Cafe-Fresh which is least expensive than all the food joints surveyed.

4) Hot Attack is mapped at centre point indicating that its give most quality vis-a-vis price.

Conclusion:

Even though it was a small exercise but its goes a long way to make us understand how we can use perceptual mapping to make our decision both at personal and business level.

We can undertake this exercise in depth including many more attributes and lot more people resulting into a comprehensive report which can be used to increase the business at the respective joints by filling in their particular deficiencies.

Submitted By

Aditya Hakim
12004
SIBM Bangalore

Perceptual Mapping

Perceptual Mapping

Perceptual mapping is also known as multidimensional sealing. It is a graphics technique and is a crucial strategic management tool that attempts to visually display the perceptions about different interrelationships of objects. It offers a unique ability to communicate the complex relationships between variables like marketplace competitors and the criteria used by buyers in making purchase decisions and recommendations, IQ of employees, various traits of people, position of a product, product line, brand, or company relative to their competition. etc.

Perceptual mapping can be used to plot the interrelationships of consumer products, industrial goods, institutions, as well as populations. Virtually any subjects that can be rated on a range of attributes can be mapped to show their relative positions in relation both to other subjects as well as to the evaluative attributes. Perceptual maps may be used for market segmentation, concept development and evaluation, and tracking changes in marketplace perceptions among other uses.

There are two methods of perceptual mapping:

Overall Similarity Method: This is used to identify how similar or how dissimilar the traits of the object are. The attributes are identified and then they are mapped. The advantage of this method is that hidden attributes are exposed. But one must have a thorough knowledge about the objects in question to be able to apply this method.

Attribute Based Method: Here the perception is mapped based on the attributes. A disadvantage of this method is one might miss out on an important attribute or may not specify an important attribute.

Perceptual mapping involves two steps: (1) data collection and (2) data analysis and presentation.

Among the various mathematical and statistical methods used to produce perceptual maps, it has been found—and published research to this effect—that multiple discriminant analysis provides the most reliable methodology. Among the reasons for this are:

1. Discriminant analysis has a close linkage between product points and attribute locations.

2. Discriminant analysis maps do not change if attributes are added that are linear combinations of those already present in the perceptual space.

3. Discriminant analysis is alone in paying attention to “between product”information, after scaling it so that “within product” differences are equal for each dimension and uncorrelated. That means that DA uses a “yardstick” to give every dimension common metric (in terms of equal unexplained variance).

4. Discriminant analysis is the most efficient method in terms of cramming into a space of low dimensionality the most information about how products differ.

5. Unlike mapping based on distances or similarities, DA make use of attribute ratings, which are easy and natural for respondents, and useful for their content even if mapping is not done with them.

Employing this methodology, respondents are never asked about similarities among products or subjects; they are asked to rate products on attributes, and similarities are inferred from differences in respondents’ ratings.

The data required for perceptual mapping thus comes from rating scales where the subjects of the map, from products to populations, are described on the basis of selected attributes. The validity of the map depends on both the overall set ofattributes and the subjects of the study as well as the subset of attributes and subjects evaluated by each respondent.

Multiple discriminant analysis uses the “F ratio” to determine attribute and product or subject location in the perceptual space. The F ratio is a ratio of the variance between ratings of different products/subjects to the variance of ratings within products/subjects. In an attribute study, these variations among ratings are generally of two types:

1. The differences between products/subjects, revealed in the difference between average ratings for different products.

2. The differences within products, revealed in the differences among respondents’ ratings of the same product. An attribute would have a higher F ratio either if its product averages were more different from one another, or if there were more agreement among respondents rating the same product.

Perceptual maps can have any number of dimensions but the most common is two dimensions. A huge change in the objective function value, which shows the error estimate of the map, means the correct picture will be obtained when the map is viewed in 3 dimension.

Name: Dibyangana Saha

Roll Number 12132

Finance Batch of 2009-11,

SIBM Bangalore.

Cluster Analysis of “Funds of Hedge Funds” Portfolio Tool

Constructing a diversified portfolio of managers in a fund of funds requires a method for determining how the different exposures complement each other.

Cluster analysis can reveal important insights into portfolio management behavior. Cluster analysis can supplement the classic tools of qualitative management interviews and this tool can, over time, help us to better understand a manager’s exposures and how a group of managers adapts to changing market opportunities.

During the last decade, funds of hedge funds have become increasingly popular with the investors who look to allocate capital to hedge funds, but do not have the resources to research, monitor, and manage a number of stand-alone hedge fund investments. Qualitative methods include in-depth interviews of the manager’s investment style and operations. Quantitative methods include the calculation of the Sharpe ratio and other performance measures, portfolio optimization, and analysis of the correlation matrix of returns.

Data and Methodology

During the last decade, funds of hedge funds have become increasingly popular with the investors who look to allocate capital to hedge funds, but do not have the resources to research, monitor, and manage a number of stand-alone hedge fund investments. Despite the fact that the fund of funds sector has been substantially affected by the current financial crisis, fund of funds’ investment in hedge funds still accounted for an estimated $606 billion of the $2.1 billion hedge fund industry at the end of the fourth quarter of 2009.

Dividing the world into specific investment strategies requires a well-established set of categories or factors that are easily identified and are known to be good predictors of the pattern of returns. In parts of the hedge fund universe these categories have been created and successfully implemented. Each fund style has three or four factors that explain a statistically significant portion of returns. These factors can be used to build replicating factor strategies at lower fee levels. Additionally, the returns of the replicating strategies can be useful to the fund of funds managers to measure the skill of hedge fund managers.

However, some strategies are heterogeneous, and organizing hedge funds into static groups may be problematic during periods of financial stress. Cluster analysis is especially useful for building peer groups and examining the change in strategies over time

Clustering Methods

Clustering methods can be divided into various groups based on their procedures for arriving at clusters and the criterion used to evaluate whether funds cluster together. The clustering methods can be divided into two main categories of hierarchal and nonhierarchical methods. Agglomerative methods and divisive methods are the two hierarchical clustering techniques. Agglomerative methods start with clusters consisting of individual managers and combine similar clusters until all funds are grouped in a single cluster. Divisive methods proceed in the opposite direction, starting with all funds in a single cluster and cleaving until each cluster contains a single fund. Nonhierarchical methods start with a fixed number of target clusters and attempt to group all funds into these target clusters.

It starts with each manager as its own cluster and then it joins two closest clusters into a new cluster with the difference between the managers defined as one minus the correlation. This process is repeated for new cluster and other cluster and process continues till we has most reflective of the data characteristics.

Each clustering method also requires a criterion for determining the similarity of funds so that funds can be clustered. Similarity measures depend on the types of input data and the goals of the clustering. For some characteristics, Euclidean distance measures can be used.

Clustering managers based on historical returns can supplement the information obtained from a qualitative review of the manager. Therefore clusters highlight which fund characteristics are most salient in understanding a strategy’s ability to diversify a portfolio.

Source:

http://post.nyssa.org/nyssa-news/2010/04/cluster-analysis-as-a-funds-of-hedge-funds-portfolio-tool.html

Submitted By:

Madhumita Das
12089
SIBM Bangalore.

Cluster analysis for identifying sub-groups and selecting potential discriminatory variables in human encephalitis

Encephalitis is an acute clinical syndrome of the central nervous system (CNS), often associated with fatal outcome or permanent damage, including cognitive and behavioral impairment, affective disorders and epileptic seizures. It is a rare disease with annual incidence ranging between 3.5 and 7.4 cases per 100,000 persons, worldwide. It is more common in children, the elderly and people with a weakened immune system (e.g. those with HIV/AIDS or cancer).

Although encephalitis has been studied extensively, there have not been any comparisons made among these studies mainly because there have not been any standard statistical methods used to describe and analyze the encephalitis data sets.

The main objective of this paper, therefore, is to perform exploratory cluster analysis using the England human encephalitis data set with the aim of achieving a better understanding of human encephalitis and generate hypotheses about etiology. In particular, we aim to: 1) identify noise variables that have little or no contribution and filter out these variables from the data; 2) identify or determine subgroups of encephalitis; 3) identify major clinical and laboratory features/characteristics associated with encephalitis, and determine which of these variables distinguish a particular cluster from others; 4) identify major risk factors associated with encephalitis.

Research

The research data is related to a prospective multicenter study where 268 patients are recruited from 24 hospitals/neurological centres in three geographical locations (South West, London, North West) across England.

Variable Selection

A total of 209 encephalitis patients and 35 variables (shown in Figure 1) are included in our initial analysis although some are filtered during the variable selection step. Complete linkage hierarchical clustering with the simple matching distance matrix is used in constructing the heatmap. However, similar results were also obtained the Euclidean and Jaccard distance matrices. The 209 by 35 dimensional matrix of encephalitis data set is displayed on the heatmap where each column represents the standardized binary measurements for a given patient. The variables that are similar for more than 80% of patients are excluded (12 variables). Variables for which a significant number of values are missing have also been removed from the analysis. The list includes Abnormal EEG, Fever, Abnormal WCC, PBChange, Abnormal Protein, Headache, Lethargy, Animal Contact, Mosquito Bite, Rash, untreated water etc.

Cluster Analysis

Clustering of our encephalitis data set using agglomerative hierarchical clustering approach revealed six major clusters. An additional small cluster, consisting of 5 encephalitis patients, was also observed. All of these 5 patients had seizures. Abnormal EEG and fever are observed in 4 (80%) of the patients. No other symptom was observed for more than one patient; moreover, most of the exposure variables are zero for these patients. The only exception is water exposure where 3 out of the 5 patients were exposed to untreated water.

The dendogram revealed marked within cluster heterogeneity and subgroups within Cluster 1. Further investigation of this cluster to identify more homogenous sub-clusters might, therefore, be of interest. With a slightly lower cut-off point, Cluster 1 can be subdivided into two sub-clusters. The first sub-group is characterized by abnormal WCC, lethargy, headache and fever; whereas, the major characteristics of the second sub-group are abnormal protein, abnormal WCC, lethargy, headache and gastro-intestinal symptoms. It is important to note that focal neurological abnormalities were observed in all of the patients in the second sub-cluster whereas only a small proportion of patients in the first sub-cluster had this symptom. Animal contact was identified as a major characteristic of sub-group 2, although only a small proportion of patients in the first sub-group were exposed to this risk factor. Another important difference between the two sub-clusters is fever.

Discussions & Conclusion

Hierarchical cluster analysis revealed six major clusters in the England encephalitis data set. An additional small cluster consisting of 5 patients was also observed in the data. It is in general assumed and is common practice to group encephalitis cases according to disease etiology. However, our results indicate that patients are clustered with respect to mainly symptom and laboratory variables rather than causal agents. In fact, disease etiology is distributed across all clusters where the majority of the patients in all the clusters are of unknown etiology. There is some intuitive sense to the clustering which deserves further exploration in relation to clusters of symptoms and laboratory variables which seem to exclude certain causes. Tuberculosis (MTB) only appears in 3 of the 6 clusters; varicella zoster virus (VZV) does not appear in clusters 4 or 5; "other viral" not in 4; acute disseminated encephalomyelitis (ADEM) is not in 5; other bacterial not in 5 or 6; and antibody-mediated encephalitis (ANT) is not found in cluster 6. Furthermore, exposure variables appear to be non-informative towards discriminating among the clusters. These similarities and/or differences with respect to symptom and laboratory measurements might, therefore, be attributed to other factors that are not included in our data set.

Among the exposure variables, however, animal contact and exposure to recent infection have only been identified as possible risk factors. Exposure to untreated water and sick person contact are also linked to encephalitis suggesting that these variables might be potential risk factors. However, further statistical analysis focused more on variable selection is required to identify potential risk factors. This might help researchers to devise new preventive measures.

The main objective of this article is mention the importance of cluster analysis is to identify the variables, sub-groups and observing the impact of the variables in human encephalitis. Through this article I have tried to portray the exact procedure followed in that process and the outcomes of the same post cluster analysis.

Source:

A research article on the same title done by Jemila S Hamid, Christopher Meaney, Natasha S Crowcroft, Julia Granerod and Joseph Beyene for on behalf of the UK Etiology of Encephalitis Study Group. Weblink: http://www.biomedcentral.com/1471-2334/10/364

Submitted By
Dipayan Kabiraj
12133

SIBM Bangalore