A Study of

The-day-of-the-week effect in
S&P CNX NIFTY
A Dissertation submitted in partial fulfillment of the requirement
for the award of M.B.A Degree of Bangalore University

By

K.BALA SHANKAR

Reg.No. 04XQCM6040

Under the Guidance of

Dr.N.Malavalli

M.P.Birla Institute of Management

Associate Bharatiya Vidya Bhavan
#43, Race Course Road
Bangalore – 560001

1
 DECLARATION

I hereby declare that this project work embodied in this

dissertation entitled “A Study of The-day-of-the-week effect in S&P
CNX NIFTY” has been carried out by me under the guidance and
supervision of Dr.N.Malavalli, M.P.B.I.M Bangalore.
I also declare that this Dissertation has not
been submitted to any University/Institution for the award of any
Degree/Diploma.

Place: Bangalore. (K. Bala Shankar)

Date:

2
 CERTIFICATE

I hereby certify that the project work

embodied in this dissertation entitled “A Study of The-day-of-the-
week effect in S&P CNX NIFTY.” has been undertaken and
completed by Mr. K.Bala Shankar under my guidance and
supervision.

I also certify that she has fulfilled all the

requirements under the covenant governing the submission of
dissertation to the Bangalore University for the award of M.B.A
Degree.

Place: Bangalore (Dr.N.Malavalli)

Date: M.P.B.I.M

3
 ACKNOWLEDGEMENT

I would like to thank my project guide and

our principle Dr.N.Malavalli,, whose contribution was

insightful and helped me, to get well acquainted to the project

intricacies.

4
 Table of Contents

Chapter 1: Abstract
Abstract 2

Chapter 2: Introduction
Need & Significance of the study 4
Objectives of the study 6
Limitations of the study 7

Chapter 3: Literature Review

Literature Review 9
Overview of Indian stock Markets 12
NSE-Overview 18

Chapter 4: Methodology
Data 23
Hypothesis 23
Statistical tools 23
Calculations Involved 24

Chapter 5: Findings & Results

Findings & Results 27

5
 Chapter 6: Analysis & Conclusion
Analysis 34
Conclusion 35

Chapter 7: Annexure
Annexure

6
Chapter 1

Abstract

7
ABSTRACT

Abstract:

The present study examines empirically the day of the week effect anomaly in
the Indian equity market for the period from 1992 to 2006using both high frequency and
end of day data for the benchmark Indian equity market index S&P CNX NIFTY. The
study mainly focuses on the returns on all the trading days are equal. In addition, there is
a perception in NSE that the returns on Monday are negative and the returns on Friday
are positive. The study is tested by ANOVA test and t-test. However, after the
introduction of the rolling settlement in 1996 the market returns are tested whether there
is any significant change. For this, the period is divided into 1992 to 1996 (before the
settlement period), 1997 to 2000, 2001 to 2004 and Jan 2005 to May 2006 (after the
settlement period).

8
 Chapter 2

Introduction

9
2. INTRODUCTION

Need and Significance of the study:

In recent years the testing for market anomalies in stock returns has become an
active field of research in empirical finance and has been receiving attention from not
only in academic journals but also in the financial press. Among the more well-known
anomalies are the size effect, the January effect and the day-of-the week effect. The
day of the week effect is a phenomenon that constitutes a form of anomaly of the
efficient capital markets theory. According to this phenomenon, the average daily return
of the market is not the same for all days of the week, as we would expect on the basis
of the efficient market theory.
January effect:
A general increase in stock prices during the month of January. This rally is
generally attributed to investors buying stocks that have dropped in price following a sell-
off at the end of December by investors seeking to create tax losses to offset any capital
gains.
The January effect is said to affect small-caps more than mid/large caps. This
historical trend, however, has been less pronounced in recent years because the
markets have adjusted for the effect. Another reason the January effect is now
considered less important is that more people are using tax-sheltered retirement plans
and therefore have no reason to sell at the end of the year for a tax loss.
Day-of-the-week effect:
The day of the week effect refers to the observation that equity returns are not
independent of the day of the week. This effect was first documented by Osborne(1962).
The last trading days of the week, particularly Friday, are characterized by positive and
substantially positive returns, while Monday, the first day of the week, and differs from
other days, even producing negative returns.

10
 Earlier studies have found the existence of the day of the week effect not only in
the USA and other developed markets but also in the emerging markets like Malaysia,
Hong Kong, Turkey. For most of the western economies, (U.S.A., U.K., Canada)
empirical results have shown that on Mondays the market has statistically significant
negative returns while on Fridays statistically significant positive returns. In other
markets such as Japan, Australia, Singapore, Turkey and France the highest negative
returns appear on Tuesdays.
The most satisfactory explanation that has been given for the negative returns on
Mondays is that usually the most unfavorable news appears during the weekends.
These unfavorable news influence the majority of the investors negatively, causing them
to sell on the following Monday. The most satisfactory explanation that has been given
for Tuesday’s negative returns are that the bad news of the weekend affecting the USA’s
market, influence negatively some markets lagged by one day.
The equity markets across many countries seem to exhibit the day-of-the-week
effect. Studies have also been conducted to identify the causes behind the patterns
observed. Institutional features of the national stock markets, such as settlement
procedures and in particular, delays between trading and settlement in the stocks,
pricing misquotes and measurement errors, specialists’ behaviour, or dividend patterns
have been put forward as the main reasons for such an effect. However none of these
reasons have been conclusively proved to be the cause of the effect. Explanations of the
day-of-the-week effect based on human nature have also been put forward to explain
the patterns observed (Jacobs and Levy, 1988). The human behaviors of disclosing
good news quickly on the weekdays and waiting for the weekend to disclose the bad
news so as to allow the market the weekend to absorb the shock, have been
explanations provided for the day-of-the-week effect.

11
Objectives of the Project

The objective of this project is to examine the day-of-the-week effect in the Indian
Stock Market. The paper in particular studies the day-of-the-week-effect with respect to
the settlement system followed. The daily closing price data on the S&P CNX NIFTY for
the period 1992-2004 has been used in the study. The first step was testing of the null
hypothesis that the mean returns on all trading days of the week are equal.

H0 = ReturnMonday = ReturnTuesday = ReturnWednesday =ReturnThursday = ReturnFriday

The null hypothesis that the means returns are equal across all trading days was
true at 5% significance level. The settlement system was changed to the weekly
settlement cycle on April 1996 in the NSE. The hypothesis was tested for the period
January 1992 to Dec 1996, for the period January 1997 to December 2000 and for the
period January 2001 to 2004 separately.

In the National Stock Exchange there is a predominant perception that the Friday
returns are lower and even negative when compared to the Monday returns. This is
because they believe that there is a selling pressure on Friday due to the weekend and
everybody is under pressure to square their positions. To test this perception the
following hypothesis was also tested. The hypothesis was tested for the period January
1992 to Dec 1996, for the period January 1997 to December 2000 and for the period
January 2001 to 2004 separately.

H0 = ReturnMonday = ReturnFriday

12
Limitations of the study:

(i) Historical Data has been used for the project study. The daily closing price
data on the S&P CNX NIFTY for the period 1992-2006 has been used in the
study.
(ii) The appropriate statistical tools ANOVA (F-Test) & t-test has been used to
test the hypothesis.

13
 Chapter 3

Literature Review

14
LITERATURE REVIEW

Literature Review:
In most developed markets such as the USA’s, the United Kingdom’s and
Canada’s, most studies, Cross (1973), Gibbons & Hess (1981), Keim & Stambaugh
(1984), Theobald and Price (1984), Jaffe & Westerfield (1985), Harris (1986), Simrlock &
Starts (1986), Board and Sutcliffe (1988), and Kohers and Kohers (1995), Tang and
Kwok (1997) for six indices [Dow Jones Industrial Average Index( US), Financial Times
Index (UK), Nikkei Average Index (Japan), Hang Seng Index (Hong Kong), FAZ General
Index (Germany) and All Ordinary Index (Australia)] and many others, have come to the
conclusion that Mondays’ average returns are negative and Fridays’ are positive. In
other words, the stock exchange market starts downwards and ends upwards. However,
in some other studies such as Condoyanni, O’Hanlon & Ward (1987), Solnik & Bousqet
(1990) in the French stock market; Athanassakos & Robinson (1994) in the Canadian
market, Jaffe & Westerfield (1985) in the stock markets of Australia and Japan, Kim
(1988) in the stock markets of Japan and Corea, Aggarwal & Rivoli (1989) in the stock
markets of Hong Kong, Singapore, Malaysia and Philippines, Ho (1990) in the stock
markets of Australia, Hong Kong, Japan, Korea, Malaysia, New Zealand, Philippines,
Singapore, Taiwan and Thailand, Wong, Hui and Chan (1992) in the markets of
Singapore, Malaysia, Hong Kong and Thailand, Dubois & Louvet (1996) in the stock
markets of Japan, Australia, Agrawal and Tandon (1994) for eighteen countries and
many others, the negative average returns are observed on Tuesdays. Also, for the
Istanbul stock exchange there were negative average returns on Tuesdays [Aydoðan
(1994), Balaban (1995), Bildik (1997) and Özmen (1997)].
On the other hand, studies on the Spanish stock market have revealed that there
is no day of the week effect, [Santemases (1986), Pena (1995) and Gardeazabal and
Regulez (2002)]. Solnik and Bousquet (1990) focused on the period 1978- 1987 and
examined the CAC Index of Paris Bourse. Their results showed strong and persistent
negative mean returns on Tuesdays. Solnik (1990) wondered whether the settlement
procedure could explain the pattern of daily returns observed in previous studies of the
Paris Bourse.

15
 Dubois and Louvet (1996) re-examined the day of the week effect for the French
stock market along with other markets such as the US, UK, German, Japanese,
Australian and Swiss markets, during the period 1969-1992 using standard statistical
approaches and moving averages. They observed that Wednesdays presented the
highest return while the day with the lowest (negative) return was Monday for all the
above markets except the Japanese and the Australian. The null hypothesis of the
equality among the mean returns of all days of the week was rejected at the 1%
confidence level. The authors concluded that probably, the different settlement systems
could account for difficulties in comparing the results internationally, but could not
explain the possible reasons for this anomaly in the US and the European markets they
examined.
If an anomaly exists in the market, the investors can take advantage of the same
and adjust their buying and selling strategies accordingly to increase their returns with
timing the market.
The day of the week effect in Indian market was examined by many researchers
(Chaudhury (1991), Poshakwala (1996), Goswami and Anshuman (2000), Choudhry
(2000), Bhattacharya, Sarkar and Mukhopadhyay (2003)). All studies except Choudhry
(2000) and Bhattacharya et al (2003) have been based on data of mid-1980s and mid-
1990s and all these studies have used conventional methods like serial autocorrelation
tests and or fitting an OLS. Choudhry (2000) examined seasonality of returns and
volatility under a unified framework but the study has a misspecification issue with
regard to conditional mean. Bhattacharya et al (2003) used GARCH framework by
incorporating the lagged returns (BSE 1001) as explanatory variables in the conditional
mean. They have used reporting and non-reporting weeks2 to study the day of the week
effect. All these studies have used end of day data.
The availability of high frequency data from NSE has opened up many avenues
of research that helps us to look closer into the market activities. The present study aims
to find the day of the week effect on India equity market using high frequency data. This
study is different in two aspects: (1) it uses the high frequency data to study the day of

16
the week effect and for the same we have to calculate the 1-minute returns and then
aggregate the same for the day to get the daily returns. This is primarily done to
understand the market dynamic observed during the whole day and to conduct a micro
analysis. The closing value that is generally available is the average of last 30 minutes of
trade and may not suitably bring out the dynamics of the market and most of the
information that happens during the day is not absorbed in the last 30 minutes of trades;
(2) the study also does a comparative analysis using the closing values to understand if
any additional valuable information can be obtained from high frequency data.
Recently there are many studies had been done on the stock market anomalies.
The research study done by Hakan Berument and Halil Kiymaz on “The Day of the
Week Effect on Stock Market Volatility: Istanbul stock exchange “ proved that the day of
the week effect is present in both volatility and return equations. While the highest and
lowest returns are observed on Wednesday and Monday, the highest and the lowest
volatility are observed on Friday and Wednesday, respectively.
There are studies had been done on the Indian stock markets. In one of the
studies done by Golaka Nath on “ day of the week effect and market efficiency –
Evidence from indian equity market using high Frequency data of national stock
exchange” proved that the study finds that before introduction of rolling settlement in
January 2002, Monday and Friday were significant days. However after the introduction
of the rolling settlement, Friday has become significant. This also indicates that Fridays,
being the last days of the weeks have become significant after rolling settlement.
Mondays were found to have higher standard deviations followed by Fridays. The
existence of market inefficiency is clear. The market inefficiency still exists and market is
yet to price the risk appropriately.

In another study done on the Indian capital markets done by Kaushik

Bhattacharya & Nityananda Sarkar on Stability of the “Day of the Week Effect in Return
and in Volatility at the Indian Capital Market” proved that in favor of significant positive
returns on non-reporting Thursday and Friday, in sharp contrast to the finding of
significant positive returns only on non-reporting Monday by OLS procedure. Separate
subperiod analyses reveal that there have been changes in daily seasonality in both
returns and volatility since the mid-1990’s at the Indian capital market, manifested in the
opposite signs and changes in the level of significance of some similar coefficients

17
across periods. These findings on the day of the week effects along with its variation
within a fortnight suggest that stock exchange regulations and the nature of interaction
between the banking sector with the capital market could possibly throw valuable
insights on the origin of the day of the week/fortnight effect in returns, while inter-
exchange arbitrage opportunities due to differences in settlement period could lead to a
seasonality in volatility.

3. Overview of the Indian Stock Market

Evolution

Indian Stock Markets are one of the oldest in Asia. Its history dates back to
nearly 200 years ago. The earliest records of security dealings in India are meagre and
obscure. The East India Company was the dominant institution in those days and
business in its loan securities used to be transacted towards the close of the eighteenth
century.

By 1830's business on corporate stocks and shares in Bank and Cotton presses
took place in Bombay. Though the trading list was broader in 1839, there were only half
a dozen brokers recognized by banks and merchants during 1840 and 1850.

The 1850's witnessed a rapid development of commercial enterprise and

brokerage business attracted many men into the field and by 1860 the number of
brokers increased into 60.

In 1860-61 the American Civil War broke out and cotton supply from United
States of Europe was stopped; thus, the 'Share Mania' in India begun. The number of
brokers increased to about 200 to 250. However, at the end of the American Civil War, in
1865, a disastrous slump began (for example, Bank of Bombay Share which had
touched Rs 2850 could only be sold at Rs. 87).

18
 At the end of the American Civil War, the brokers who thrived out of Civil War in
1874, found a place in a street (now appropriately called as Dalal Street) where they
would conveniently assemble and transact business. In 1887, they formally established
in Bombay, the "Native Share and Stock Brokers' Association" (which is alternatively
known as " The Stock Exchange "). In 1895, the Stock Exchange acquired a premise in
the same street and it was inaugurated in 1899. Thus, the Stock Exchange at Bombay
was consolidated.

Other leading cities in stock market operations

Ahmedabad gained importance next to Bombay with respect to cotton textile

industry. After 1880, many mills originated from Ahmedabad and rapidly forged ahead.
As new mills were floated, the need for a Stock Exchange at Ahmedabad was realised
and in 1894 the brokers formed "The Ahmedabad Share and Stock Brokers'
Association".

What the cotton textile industry was to Bombay and Ahmedabad, the jute
industry was to Calcutta. Also tea and coal industries were the other major industrial
groups in Calcutta. After the Share Mania in 1861-65, in the 1870's there was a sharp
boom in jute shares, which was followed by a boom in tea shares in the 1880's and
1890's; and a coal boom between 1904 and 1908. On June 1908, some leading brokers
formed "The Calcutta Stock Exchange Association".

In the beginning of the twentieth century, the industrial revolution was on the way
in India with the Swadeshi Movement; and with the inauguration of the Tata Iron and
Steel Company Limited in 1907, an important stage in industrial advancement under
Indian enterprise was reached.

Indian cotton and jute textiles, steel, sugar, paper and flour mills and all
companies generally enjoyed phenomenal prosperity, due to the First World War.

19
 In 1920, the then demure city of Madras had the maiden thrill of a stock
exchange functioning in its midst, under the name and style of "The Madras Stock
Exchange" with 100 members. However, when boom faded, the number of members
stood reduced from 100 to 3, by 1923, and so it went out of existence.

In 1935, the stock market activity improved, especially in South India where there was a
rapid increase in the number of textile mills and many plantation companies were
floated. In 1937, a stock exchange was once again organized in Madras - Madras Stock
Exchange Association (Pvt) Limited. (In 1957 the name was changed to Madras Stock
Exchange Limited).

Lahore Stock Exchange was formed in 1934 and it had a brief life. It was merged with
the Punjab Stock Exchange Limited, which was incorporated in 1936.

Indian Stock Exchanges - An Umbrella Growth

The Second World War broke out in 1939. It gave a sharp boom which was
followed by a slump. But, in 1943, the situation changed radically, when India was fully
mobilized as a supply base.

On account of the restrictive controls on cotton, bullion, seeds and other

commodities, those dealing in them found in the stock market as the only outlet for their
activities. They were anxious to join the trade and their number was swelled by
numerous others. Many new associations were constituted for the purpose and Stock
Exchanges in all parts of the country were floated.

The Uttar Pradesh Stock Exchange Limited (1940), Nagpur Stock Exchange
Limited (1940) and Hyderabad Stock Exchange Limited (1944) were incorporated.

In Delhi two stock exchanges - Delhi Stock and Share Brokers' Association Limited and
the Delhi Stocks and Shares Exchange Limited - were floated and later in June 1947,
amalgamated into the Delhi Stock Exchnage Association Limited.

20
Post-independence Scenario

Most of the exchanges suffered almost a total eclipse during depression. Lahore
Exchange was closed during partition of the country and later migrated to Delhi and
merged with Delhi Stock Exchange.

Bangalore Stock Exchange Limited was registered in 1957 and recognized in 1963.

Most of the other exchanges languished till 1957 when they applied to the
Central Government for recognition under the Securities Contracts (Regulation) Act,
1956. Only Bombay, Calcutta, Madras, Ahmedabad, Delhi, Hyderabad and Indore, the
well established exchanges, were recognized under the Act. Some of the members of
the other Associations were required to be admitted by the recognized stock exchanges
on a concessional basis, but acting on the principle of unitary control, all these pseudo
stock exchanges were refused recognition by the Government of India and they
thereupon ceased to function.

Thus, during early sixties there were eight recognized stock exchanges in India
(mentioned above). The number virtually remained unchanged, for nearly two decades.
During eighties, however, many stock exchanges were established: Cochin Stock
Exchange (1980), Uttar Pradesh Stock Exchange Association Limited (at Kanpur, 1982),
and Pune Stock Exchange Limited (1982), Ludhiana Stock Exchange Association
Limited (1983), Gauhati Stock Exchange Limited (1984), Kanara Stock Exchange
Limited (at Mangalore, 1985), Magadh Stock Exchange Association (at Patna, 1986),
Jaipur Stock Exchange Limited (1989), Bhubaneswar Stock Exchange Association
Limited (1989), Saurashtra Kutch Stock Exchange Limited (at Rajkot, 1989), Vadodara
Stock Exchange Limited (at Baroda, 1990) and recently established exchanges -
Coimbatore and Meerut. Thus, at present, there are totally twenty one recognized stock
exchanges in India excluding the Over The Counter Exchange of India Limited (OTCEI)
and the National Stock Exchange of India Limited (NSEIL).

21
 The Table given below portrays the overall growth pattern of Indian stock
markets since independence. It is quite evident from the Table that Indian stock markets
have not only grown just in number of exchanges, but also in number of listed
companies and in capital of listed companies. The remarkable growth after 1985 can be
clearly seen from the Table, and this was due to the favouring government policies
towards security market industry.

Source : Various issues of the Stock Exchange Official Directory, Vol.2 (9) (iii),
Bombay Stock Exchange, Bombay.

Trading Pattern of the Indian Stock Market

Trading in Indian stock exchanges are limited to listed securities of public limited
companies. They are broadly divided into two categories, namely, specified securities
(forward list) and non-specified securities (cash list). Equity shares of dividend paying,
growth-oriented companies with a paid-up capital of atleast Rs.50 million and a market
capitalization of atleast Rs.100 million and having more than 20,000 shareholders are,
normally, put in the specified group and the balance in non-specified group.

Two types of transactions can be carried out on the Indian stock exchanges: (a)
spot delivery transactions "for delivery and payment within the time or on the date
stipulated when entering into the contract which shall not be more than 14 days following
the date of the contract" : and (b) forward transactions "delivery and payment can be
extended by further period of 14 days each so that the overall period does not exceed 90
days from the date of the contract". The latter is permitted only in the case of specified
shares. The brokers who carry over the outstandings pay carry over charges (cantango
or backwardation) which are usually determined by the rates of interest prevailing.

A member broker in an Indian stock exchange can act as an agent, buy and sell
securities for his clients on a commission basis and also can act as a trader or dealer as
a principal, buy and sell securities on his own account and risk, in contrast with the
practice prevailing on New York and London Stock Exchanges, where a member can act
as a jobber or a broker only.

22
 The nature of trading on Indian Stock Exchanges are that of age old conventional
style of face-to-face trading with bids and offers being made by open outcry. However,
there is a great amount of effort to modernize the Indian stock exchanges in the very
recent times.

Bombay Stock Exchange (BSE) and National Stock Exchange of India Ltd.
(NSE) are the two primary exchanges in India. In addition there are 22 Regional Stock
Exchanges. However BSE and NSE have established themselves as the two main
exchanges and account for about 80% of the volume traded in Indian equity markets.
Approximately NSE and BSE are equal in size in terms of daily volume traded. NSE has
about 1500 shares which are traded and has a total market capitalisation of around Rs.
9,21,500 Crores (Rs. 9215-bln). BSE has over 6000 stocks traded and has a total
market capitalisation of around Rs. 9,68,000 Crores (Rs.9680-bln). Most key stocks are
available on both exchanges and hence the investor could buy them on either exchange.
Both exchanges have a different settlement cycles. The primary index of BSE is BSE
Sensex, which comprises of 30 Stocks. NSE has its own S& P NSE 50 Index (Nifty)
which comprises of fifty stocks. Both these indexes are calculated on the basis of market
capitalisation and contain the most highly traded shares from the key sectors. Daily
volume on both exchanges is approximately Rs. 4500 Crores each. (Rs. 45-bln.) The
key regulator governing Stock Exchanges, Brokers, Depositories, Depository
participants, Mutual Funds, FIIs and other participants in Indian secondary and primary
market is Securities and Exchange Board of India (SEBI) Ltd.

23
National Stock Exchange (NSE)

With the liberalization of the Indian economy, it was found inevitable to lift the
Indian stock market trading system on par with the international standards. On the basis
of the recommendations of high powered Pherwani Committee, the National Stock
Exchange was incorporated in 1992 by Industrial Development Bank of India, Industrial
Credit and Investment Corporation of India, Industrial Finance Corporation of India, all
Insurance Corporations, selected commercial banks and others.

Trading at NSE can be classified under two broad categories:

(a) Wholesale debt market and

(b) Capital market.

Wholesale debt market operations are similar to money market operations - institutions
and corporate bodies enter into high value transactions in financial instruments such as
government securities, treasury bills, public sector unit bonds, commercial paper,
certificate of deposit, etc.

There are two kinds of players in NSE:

(a) trading members and

(b) participants.

Recognized members of NSE are called trading members who trade on behalf of
themselves and their clients. Participants include trading members and large players like
banks who take direct settlement responsibility.

Trading at NSE takes place through a fully automated screen-based trading mechanism
which adopts the principle of an order-driven market. Trading members can stay at their
offices and execute the trading, since they are linked through a communication network.
The prices at which the buyer and seller are willing to transact will appear on the screen.

24
When the prices match the transaction will be completed and a confirmation slip will be
printed at the office of the trading member.

NSE has several advantages over the traditional trading exchanges. They are as
follows:

 NSE brings an integrated stock market trading network across the nation.

 Investors can trade at the same price from anywhere in the country since inter-
market operations are streamlined coupled with the countrywide access to the
securities.

 Delays in communication, late payments and the malpractice’s prevailing in the

traditional trading mechanism can be done away with greater operational
efficiency and informational transparency in the stock market operations, with the
support of total computerized network.

Unless stock markets provide professionalised service, small investors and foreign
investors will not be interested in capital market operations. And capital market being
one of the major source of long-term finance for industrial projects, India cannot afford to
damage the capital market path. In this regard NSE gains vital importance in the Indian
capital market system.

Settlement Cycle

Settlement refers to the process whereby payment is made by all those who
have made purchases and shares are delivered by all those who have made sales. The
exchange ensures that buyers, who have paid for the shares purchased, receive the
shares. Similarly sellers who have given delivery of shares to the exchange receive
payment for the same. The entire process of settlement of shares and money is
managed by stock exchanges (SEs) through Clearing House (CH) which are entities
formed specifically to ensure that the process of settlement takes place smoothly.

25
 Settlement Cycle refers to a calendar according to which all purchase and sale
transactions done within the dates of the settlement cycle are settled on a net basis.
NSE and BSE currently follow their own weekly settlement cycles. SEBI has introduced
a rolling settlement cycle from Jan 12, 2000. Currently 43 stocks are traded in rolling
settlement cycles. All other stocks are traded in the weekly settlement cycles of SEs.
SEBI plans to add more and more stocks to rolling settlement cycle by moving them out
of the weekly settlement cycle. A brief description of various settlement cycles is given
below:

NSE Settlement Cycle

Before the settlement period introduced the BSE and the NSE followed two
different settlement weeks. While the BSE followed a Monday to Friday cycle, for the
NSE, it is from Wednesday to next Tuesday.

This has provided market players, mainly speculators, an opportunity to shift their
position from one exchange to another depending on the end of settlement week. This
has also afforded an arbitrage opportunity for the big operators. All this would become
passed with the introduction of the uniform settlement in the two exchanges, which have
virtually decimated the smaller regional exchanges after trading went online.

The NSE has stated that SEBI has listed 414 securities, which are included in the
ALBM/BLESS/MCFS and BSE 200 list, for trading only in the compulsory rolling
settlement. But in the NSE this covered only 301 securities as the rest of them were
either not listed or not traded on the NSE.

In the compulsory rolling settlement, traders/buyers cannot carry forward their

position to the next day. They will have to either square off their position within that day
or take delivery of scrips/receive payment on the due date.

The remaining listed and permitted securities would be available in both the
rolling and account period settlement

26
Rolling Settlement Cycle:

The Exchange has commenced trading in the Dematerialised (Demat) segment

with effect from December 29, 1997 where there is no physical delivery of securities as
in the physical segment. Trading in the Demat segment is on a Rolling Settlement basis
(T+5) where T stands for Trade Day. The pay-in and pay-out for the transactions in this
segment are both conducted on a single day. The Pay-in & Pay-out for transactions
executed on Monday is conducted on the following Monday, i.e., corresponding day in
the following week. Auction session for shortages in demat segment is conducted on
BOLT on the day after pay-in/pay-out. The pay-in / pay-out (money part) takes place
through computerised posting of debits and credits in the members’ bank accounts as in
the case of physical segment

Hence unlike a BSE or NSE weekly settlement cycle where one buys or sells
shares on the beginning of the settlement cycle and can decide till the end of the
settlement cycle whether to give delivery or make payment or square of the transaction
by covering. In a rolling settlement cycle, decision has to be made by end of trading on
the same day. Rolling settlement cycles have been recently introduced on both
exchanges form January 12, 2000. To start with only 43 shares will be traded on the
Rolling Settlement Cycle.

27
 Chapter 4

Methodology

28
METHODOLOGY:

Data:

The daily closing price data on the S&P CNX NIFTY for the period 1992-2004 has
been used in the study. S&P CNX Nifty is a well diversified 50 stock index accounting for
25 sectors of the economy. It is used for a variety of purposes such as benchmarking
fund portfolios, index based derivatives and index funds.

Hypothesis:

To the day-of-the-week present in the NSE, the first step was testing of the null
hypothesis that the mean returns on all trading days of the week are equal.

H0 = ReturnMonday = ReturnTuesday = ReturnWednesday =ReturnThursday = ReturnFriday

To test the predominant perception that the Friday returns are lower and even
negative when compared to the Monday returns, the following Hypothesis is taken.

H0 = ReturnMonday = ReturnFriday

Statistical tests used:

To the chosen hypothesis, stastical tools have been used.

 ANOVA ( F-Test)
 t-test

Both the tests are performed to see whether there are any significant deviations
from the means. The ANOVA test is used when there is 23 or more groups are involved
and the t-test is used to test the significant deviation in two means.

29
ANOVA ( F-Test):

Analysis of Variance (ANOVA) is statistical method used to compare two or more

means. It may seem odd that the technique is called "Analysis of Variance" rather than
"Analysis of Means”. ANOVA is used to test general rather than specific differences
among means.

Analysis of Variance (ANOVA) allows us to extend this to more than two

populations or measurements (treatments). That is, we can test the following:

 Are all the means from more than two populations equal?
 Are all the means from more than two treatments on one population equal?

T-test:

A t-test is an inferential test that determines if there is a significant difference

between the means of two data sets. In other words, a t-test decides if the two data sets
come from the same population or from different populations).

Calculations Involved:

From the collected data, the returns are calculated as given below

Return = ln ( Vt/Vt-1)

Where

Vt =the closing value of the index on day t

Vt-1 = the closing value of the index on the previous day

The hypothesis was tested using the F test at 0.05 level of significance. The
hypothesis was tested on different periods of data to check whether the day-of-the-week
effect varies with time. The periods were chosen such that they reflected periods where
different settlement systems were followed. The settlement system was changed to the

30
weekly settlement cycle on April 1996 in the NSE. The hypothesis was tested for the
period January 1992 to Dec 1996, January 1997 to Dec 2000and for the period January
2001 to Dec 2004 separately. The idea was to check if the change in the settlement
system induced any change in the mean returns.

31
 Chapter 5

Findings & Results

32
Results and Findings:

(i) Hypothesis Tested: Mean returns are equal across all trading days of the
week

H0=ReturnMonday = ReturnTuesday = ReturnWednesday =ReturnThursday = ReturnFriday

Findings:

Hypothesis
period Fcal Ftable Probability
Result
Null hypothesis
Jan 1992 to Dec 2005 1.725 2.37 14%
can’t be rejected
Null hypothesis
Jan 1992 to Dec 1996 1.589 2.39 18%
can’t be rejected
Null hypothesis
Jan 1997 to Dec 2000 1.446 2.39 22%
can’t be rejected
Null hypothesis
Jan 2001 to Dec 2004 0.1262 2.22 25%
can’t be rejected
Null hypothesis
Jan 2005 to May 2005 0.3728 2.40 83%
can’t be rejected

Means for the various periods:

Jan 1992 Jan 1992 Jan 1997 Jan 2001 Jan 2005
Days to Dec to Dec to Dec to Dec to May
2004 1996 2000 2004 2006

Monday .00256 0.00186 0.0042 0.002 0.006

Tuesday 0.00161 0.003 0.001 0.0013 0.005

Wednesday 0.00106 0.0031 0.001 0.0014 0.004

Thursday 0.00161 0.00.31 0.0037 0.001 0.0067

Friday 0.001 0.0019 0.001 0.001 0.0072

33
Results:

Results of the hypothesis that the mean returns on all trading days of the week
are equal. The null hypothesis that the means returns are equal across all trading days
is accepted at 5% significance level.

The null hypothesis as can be seen from above Table was accepted in all the
four cases shown. The weekly settlement system came into being in April 1996, and the
very fact that the hypothesis is proven for all periods show that the settlement system did
not produce any day-of-the-week effect in the Indian stock market. More important and
startling is the conclusion that Indian stock markets are indeed efficient as far as the
day-of-the-week effect is concerned.

For the period chosen from 1992 to 2006, the returns from all the trading days are
equal from the data taken and proved from the test. Therefore, in the Indian stock market
context the day-of-the-week effect is not present.All the results for the four cases are
shown in the following Exhibits.

Exhibits 1: Jan 1992 to May 2006

Source Of Degrees of
Sum of Square Mean Square
Variation Freedom

Between (column) 0.002354 4 0.00059

Within(Error) 0.6361 1866 0.00034

Total 0.6384 1870

34
Exhibits 2: Jan 1992 to Dec 1996

Source Of Degrees of
Sum of Square Mean Square
Variation Freedom

Between (column) 0.0029 4 0.00072

Within(Error) 0.2962 650 0.00046

Total 0.2991 654

Exhibits 3: Jan 1997 to Dec 2000

Source Of Degrees of
Sum of Square Mean Square
Variation Freedom

Between (column) 0.00178 4 0.00044

Within(Error) 0.1715 557 0.00031

Total 0.1733 561

Exhibits 4: Jan 2001 to Dec 2004

Source Of Degrees of
Sum of Square Mean Square
Variation Freedom
Between
0.00014 5 0.00035
(column)
Within(Error) 0.1563 570 0.00027

Total 0.1565 574

35
Exhibits 4: Jan 2001 to May 2006

Source Of Degrees of
Sum of Square Mean Square
Variation Freedom
Between
2.1010E-03 4 5.2526E-04
(column)
Within(Error) 0.4767 342 1.3939E-04

Total 0.4788 346

36
(ii) Hypothesis is tested:

H0 = ReturnMonday = ReturnFriday

The above hypothesis is tested to evaluate the returns on Monday & Friday is equal.

period p á Hypothesis Result

Null hypothesis
Jan 1992 to May 2006 0.5 0.1
can’t be rejected
Null hypothesis
Jan 1992 to Dec 1996 0.124 0.1
can’t be rejected
Null hypothesis
Jan 1997 to Dec 2000 0.989 0.1
can’t be rejected
Null hypothesis
Jan 2001 to Dec 2004 0.69 0.1
can’t be rejected
Null hypothesis
Jan 2005 to may 2006 0.81 0.1
can’t be rejected

The above table shows Results of the hypothesis that the mean returns are equal
across Friday and Monday. The results show that the null hypothesis is true for the
complete period from January 1992 to May 2006 hypothesis is tested at .05%
significance.

As the per the t-test, in all the cases p > á , the null hypothesis cannot be rejected.

For the period chosen from 1992 to 2006, the returns on the Mondays and the
Fridays are equal from the data taken and proved from the test. Therefore, in the Indian
stock market context the day-of-the-week effect is not present.

37
Results from Jan 1992 to Dec 1996:

Mean of X 0.001858
Mean of Y -0.001945
Standard Deviation of X 0.025830
Standard Deviation of Y 0.022389
Large Sample Standard Deviation 0.003285

t-statistic 1.157599
degrees-of-freedom 163.958525
p-value 0.124356

Results from Jan 1997 to Dec 2000:

Mean of X -0.001087
Mean of Y 0.004294
Standard Deviation of X 0.015947
Standard Deviation of Y 0.018368
Large Sample Standard Deviation 0.002323

t-statistic -2.316816
degrees-of-freedom 208.770327
p-value 0.989258

Results from Jan 2000 to Dec 2004:

Mean of X 0.000744
Mean of Y 0.001507
Standard Deviation of X 0.014555
Standard Deviation of Y 0.014840
Large Sample Standard Deviation 0.001938

38
 t-statistic -0.393531
degrees-of-freedom 227.913936
p-value 0.694295

Results from Jan 1992 to Dec 2004:

Mean of X 0.002560
Mean of Y -0.000878
Standard Deviation of X 0.019685
Standard Deviation of Y 0.018423
Large Sample Standard Deviation 0.001462

t-statistic 2.350966
degrees-of-freedom 641.521269
Results from Jan 1992 p-value 0.009513 to May 2006:

Mean of X 0.005898
Mean of Y 0.007158
Standard Deviation of X 0.030550
Standard Deviation of Y 0.031872
Large Sample Standard Deviation 0.005364

t-statistic -0.234964
degrees-of-freedom 131.732143
p-value 0.814601

39
 Chapter 6

Analysis & Conclusions

40
Analysis:

The study focuses on the returns on all the trading days are equal or not. In
addition, to test the perception of NSE that the returns on Mondays are positive and
Fridays are negative.

The null hypothesis as can be seen from above Table was accepted in all the
four cases shown. The weekly settlement system came into being in April 1996, and the
very fact that the hypothesis is proven for all periods show that the settlement system did
not produce any day-of-the-week effect in the Indian stock market. More important and
startling is the conclusion that Indian stock markets are indeed efficient as far as the
day-of-the-week effect is concerned.

For the period chosen from 1992 to 2004, the returns from all the trading days
are equal from the data taken and proved from the test. Therefore, in the Indian stock
market context the day-of-the-week effect is not present The results also showed that
the perception of the NSE of non-significant from the data collected.

Results of the hypothesis that the mean returns are equal across Friday and
Monday. The results show that the null hypothesis is true for the complete period from
January 1992 to Dec2004. This hypothesis was tested at the significance of 5%.

For the period chosen from 1992 to 2004, the returns on the Mondays and the
Fridays are equal from the data taken and proved from the test. Therefore, in the Indian
stock market context the day-of-the-week effect is not present.

41
Conclusion:

The study has proved that there is day-of-the-week effect is not present in the of
market index from 1992-2004. Nevertheless, the data is only represents certain period
from 1992 to 2004 and only that S&P CNX NIFTY of market index. In addition to that
there are different periods are chosen to test during that particular period is there any
day-of-the-week effect persent.this is because to test, roll settlement system that started
in 1996 has any effect on the returns.
The study had proved that day- of- the- week effect is not present during the
January 1992 and December 2004 periods by using S&P CNX NIFTY. After the
introduction of the rolling settlement in 1996 the market returns for 1992 to 1996 (before
the settlement period), 1997 to 2000 and 2001 to 2004 (after the introduction of the
settlement periods) are same.

So the research study showed that there is no day-of-the-week effect present in

the S&P CNX NIFTY of market index.

42
 Chapter 7

ANNEXURE

43
Annexure:

 www.nseindia.com
 www.physics.csbsju.edu/stats/anova.html
 http://www.socialresearchmethods.net/kb/stat_t.htm

44