project report kevin

7/31/2019 Project Report Kevin

1/16

A Summer Project Proposal for

PGDM

By

Kevin Dinesh GalaBatch 2011-2013

PGDM B - 119

Under the guidance of

Date27st June, 2012

1


2/16

Abstract

By

Kevin Dinesh Gala

This project has been a great learning experience for me; it has helped me to learn about

portfolio risk measurement using multi factor model and the impact of intra industry and inter

industry correlation.

The projects included computing intra industry correlations and inter industry

correlations which are used for computing factor loading parameter. The project was to select a

representative set of listed companies which represent the industries correctly in model.

2


3/16

Contents

Abstract............................................................................................................................................ 2

Introduction.................................................................................................................................. 4

3


4/16

Introduction

4


5/16

Multi factor model for portfolio risk index caculation

5


6/16

Introduction of the company

ICICI Bank is India's second-largest bank with total assets of Rs. 4,062.34 billion

(US$ 91 billion) at March 31, 2011 and profit after tax Rs. 51.51 billion (US$ 1,155

million) for the year ended March 31, 2011. The Bank has a network of 2,752 branches and

9,225 ATMs in India, and has a presence in 19 countries, including India.

ICICI Bank offers a wide range of banking products and financial services to

corporate and retail customers through a variety of delivery channels and through its

specialised subsidiaries in the areas of investment banking, life and non-life insurance,venture capital and asset management.

The Bank currently has subsidiaries in the United Kingdom, Russia and Canada,

branches in United States, Singapore, Bahrain, Hong Kong, Sri Lanka, Qatar and Dubai

International Finance Centre and representative offices in United Arab Emirates, China,

South Africa, Bangladesh, Thailand, Malaysia and Indonesia. Our UK subsidiary has

established branches in Belgium and Germany.

ICICI Bank's equity shares are listed in India on Bombay Stock Exchange and the

National Stock Exchange of India Limited and its American Depositary Receipts (ADRs)

are listed on the New York Stock Exchange (NYSE).

6


7/16

Profile of the company

Name of the company: ICICI Bank

Year of Establishment:

Headquarter :

Nature of Business: Banking Service

Services: Banking products, financial service, investment banking, life and non-life insurance, venture capitaland asset management

Number of Employees:

Website: www.icicibank.com

Slogan: Khayaal aapka.

Vision:

.

Mission:

.

7


8/16

Equity Liquidity

The model classifies Indian companies under nineteen different industries. Each industry is

represented by group of companies. Selection of companies for representing a that particular

industry is based on its liquidity. Here liquidity refers to equity liquidity and not financial ratios

like current ratio and quick ratio. Equity liquidity is measured on following parameters:-

1. As liquidity for an asset is determined by how easily it can be converted to cash. In similar

manner liquidity for equity is determined by how easily shares of stock can be converted to cash.

The market for a stock is said to be liquid if the shares can be rapidly sold and the act of selling

has little impact on the stock's price. Generally, this translates to where the shares are traded and

the level of interest that investors have in the company. Company stock traded on the major

exchanges can usually be considered liquid. Often, approximately 1% of the float trades hands

daily, indicating a high degree of interest in the stock.

2. Equity liquidity can also be determined by the bid ask spread for a companys stock. Bid price

is the price that buyer of stock quotes and ask price is the price at which a seller is willing to sell

the stock of company. For liquid stocks the bid ask spread is much less than 1% of the price. For

illiquid stocks, the spread can be much larger amounting to a few percent of the trading price.

The model uses equity correlation as a proxy for asset correlation

8


9/16

Exercise 1

An analysis was done for each industry present in model and for each industry present in portfolio of

ICICI bank that was used to determine the risk index. The findings of the analysis were as follows:-

1. The portfolio included listed and non listed companies under each industry. So first part was to

find the listed companies under each industry in the portfolio and flag them. This was done in order to

find the characteristic of each industry as to find the number of listed and non listed companies for each

industry in the portfolio.

2. The companies present in the portfolio were compared with the companies present in the model

under each industry to compute correlations. This was to find out the number of same companies

present in the portfolio and also present in the model for each industry. Using this information it was

found the number of listed companies present in the portfolio and not present in the model for each

industry. Whether the listed companies present in the portfolio should be used in the model to represent

that particular industry to compute correlations.

This exercise helped in finding the mapping between companies present in the portfolio and the

companies present in the model. The excel sheet for this exercise is attached below.

Exercise 2

For each industry present in model a list of listed companies along with their market capital was

collected using Bloomberg as data source and following were the findings:-

9


10/16

1. For each industry the total market capital was calculated and for each company under that

industry percentage market capital of total capital was calculated.

2. The companies present in the model under a particular industry were flagged and the total

percentage of market capital was calculated. This was done was all industries.

3. For each industry, total percentage of market capital for top 10 companies based on market

capital was computed.

This exercise was used to compare the total market capital of top 10 companies with that of companies

present in the model for each industry. As a conservative approach the percentage of market capital of

companies under a industry should be greater that top 10 companies for each industry. With the help of

this exercise it was found that for some of the industries the model did not include the top 10

companies. The excel sheet for this exercise is attached below.

Exercise 3

To find the criteria for selecting the companies to represent a particular industry used in model a

exercise was conducted for Utilities, Chemical and Fertilizers, Pharmaceuticals and Biotechnology

10


11/16

industries. Following were the finding:-

1. The model included companies having market capital greater than 0.1% of total market capital

of that industry.

2. Among the companies selected from above criteria only those companies were selected having

average bid ask spread less than 0.5.

From the above findings it was concluded that for an industry, companies having large market capital

and highly liquid in terms of equity liquidity (as explained in equity liquidity section ) are affected by

changes in economics and market characteristics and are useful to represent the behavior of a particular

industry to changes of economics and market characteristics.

Exercise 4

Some companies have a diversified profile and have presence in various industries. Classifying those

companies under a particular industry was done based on method as stated below:-

1. The company description was collected for such companies from Bloomberg data source. From

this description it was identified the various industries the company has its presence. For some of thecompanies based on its description it was identified to which industry they should be classified.

2. For remaining companies, from which business segment major part of revenue is generated was

found and the company was classified under that industry.

This method helped in classifying companies having presence in various industries under particular

industry. The sheet having examples of such companies is attached below.

Exercise 5

The model did not include NBFC (Non banking financial companies) industry for computing

correlations. So list of companies use to represent NBFC industry was to be formed to use in model to

compute the correlations. Following method was use:-

11


12/16

1. A list of companies under NBFC was collected from Bloomberg used as data source. Based on

method explained in exercise 3 a list of companies was selected for representing NBFC industry.

2. NBFC as industry includes companies such as rating companies, exchanges, holding

companies, Gold ETF companies. After discussion with industry analysis team of ICICI bank such

companies were not included in NBFC industry as these companies does not offer financial services.

So such companies were excluded from the list.

Using above method a list of companies representing NBFC (Non banking financial companies)

industry was formed.

Exercise 6

The model did not include Retail and Gems/Jewelry industry for computing correlations. So list of

companies use to represent Retail and Gems/Jewelry industry was to be formed to use in model to

compute the correlations. Following method was use:-

1. According to ICICI bank their definition of retail industry includes companies doing business

in apparel retail only. Based on this definition four companies were selected from retail industry. Thelist of these companies is attached below.

2. For Gems/Jewelry industry ICICI bank defines it as including those companies involved in

retail business of gems and jewelry. Based on this definition of gems/jewelry industry five companies

were selected under this industry. The list of these companies is attached below.

Using industry definition for retail and gems/jewelry industry a list of companies under this industry

was formed having nine companies in all.

Exercise 7

After forming list of companies to represent NBFC and Retail and Gems/Jewelry industry equity data

12


13/16

(stock price) was collected for those companies from Nov 2010 to March 2012 from Bloomberg. Using

these equity data correlations was computed. The model converted equity data (stock price) to returns

data using price2ret function of matlab. The working of this function is, current day closing price of

stock is converted to returns by diving current day closing price of stock by previous day closing of

stock price and taking natural log of the ratio. For companies under NBFC and Retail and

Gems/jewelry industries the equity data (stock price) was converted to returns data as done by price2ret

function and correlations were computed. Following was the method for computation of intra industry

and inter industry:-

Intra industry correlation

Intra industry correlation were computed by combining each company present in the NBFC industry

with other companies present in that industry. The correlations were computed using moving average

of one month with three month look back period. There are 32 companies present in NBFC industry so

in all there were 496 (31*16=496) combinations for computing correlations using moving average for

one month with three months look back period. The average of these combinations was the intra

industry correlation for NBFC industry.

Similar steps were followed to compute intra industry correlation was Retail and Gems/jewelry

industry. The intra industry correlation was computed for following periods:-

Period

One month movingaverage period

Three month lookback period

Mar-12 1st March 2012 to 31st March 201231st March 2012 to 1st Dec2011

Dec-11 1st Dec 2011 to 31st Dec 2011 31st Dec 2011 to 1st Sep 2011

Sep-11 1st Sep 2011 to 3oth Sep 2011 3oth Sep 2011 to 1st June 2011

Jun-11 1st June 2011 to 3oth June 20113oth June 2011 to 1st March2011

Mar-11 1st March 2011 to 31st March 201131st March 2011 to 1st Dec2010

Following table shows the value of intra industry correlation for NBFC and Retail and Gems/Jewelry

industry for above mentioned periods:-

13


14/16

Inter industry correlation

For inter industry correlation NBFC and Tech Hardware and Equipment was used. The equity data was

collected for companies under NBFC and Tech Hardware and Equipment and were converted to

returns data as done by price2ret function of matlab. Inter industry correlation were computed by

combining each company present in the NBFC industry with each company under Tech Hardware and

Equipment industry. The correlations were computed using moving average of one month with three

month look back period. There are 32 companies present in NBFC industry and 16 companies presentunder Tech Hardware and Equipment industry so in all there were 512(32*16=512) combinations for

computing correlations using moving average for one month with three months look back period. The

average of these combinations was the inter industry correlation for NBFC and Tech Hardware and

Equipment industry.

Similar steps were followed to compute inter industry correlation between Retail and Gems/jewelry

industry and Tech Hardware and Equipment industry. The inter industry correlation was computed for

periods as stated in above table. Following table shows the value of inter industry correlations for

NBFC and Retail and Gems/Jewelry industry with Tech Hardware and Equipment industry for above

mentioned periods:-

After this exercise intra industry correlation for NBFC and Retail and Gems/Jewelry industry and inter

industry correlations between NBFC and Retail and Gems/Jewelry were computed and this was

14


15/16


16/16

16

project report kevin

Documents