INDIAN STOCK MARKET EFFICIENCY AN ANALYSIS

Transcription

1 CHAPTER V INDIAN STOCK MARKET EFFICIENCY AN ANALYSIS The Indian stock market is considered to be one of the earliest in Asia and is regarded as the barometer of the health of the Indian economy. In line with the global trend, reforms of the Indian stock market also started with the establishment of Securities and Exchange Board of India (SEBI). With the establishment of SEBI and technological advancement Indian stock market has now reached the global standards. The major indicators of stock market development show that significant development has taken place in the Indian stock market during the post-reform period. The adoption of international quality in trading and settlement mechanisms and the reduction of transaction costs, removal of barriers to the international equity investment, better allocation and mobilization of resources have made the investors both domestic and foreign to be more optimistic which in turn evidenced a considerable growth in market volume and liquidity. Together, all these market features infer better market efficiency in Indian stock market.

2 Efficient Market Hypothesis Efficient Market Hypothesis is an investment theory which states that it is impossible to beat the market because market efficiency causes exiting share prices to always incorporate and reflect all relevant information. Stocks are always traded at their fair value on stock exchanges and so the scope of residual returns either by purchasing undervalued stocks or by selling stocks for inflated prices is impossible.in an efficient market, prices fully and instantaneously reflect all available information. Ever since Fama (1965) propounded his famous Efficient Market Hypothesis (EMH), a number of empirical studies have been conducted to test its validity, both in developed markets and as well as in emerging markets. The contradictory nature of the results and the change in the current market scenario encouraged the researcher to conduct a research in the market efficiency of Indian Stock Market. Market Efficiency can be explained in three related concepts: Operational Efficiency, Allocation Efficiency and Informational Efficiency. Operational efficiency ensures that all transactions are completed on time, with maximum accuracy and at least cost. Allocation efficiency talks about capital flow to the projects with highest possible risk-adjusted returns whereas Informational efficiency ensures that market price of a security fully reflect all information which is affecting the pricing of security.

3 Efficient Market Hypothesis mainly discusses about informational efficiency and states that markets are efficient if the prices of securities fully reflect all available information. Again the theory talks about three forms of efficiency: 86 Weak Form Efficiency Semi-strong Form efficiency Strong Form One cannot beat the market by using historical information on prices of securities if the market is said to be weak form efficient. Semi-strong efficiency implies that the current prices of stocks of various companies reflects not only the information on historical prices but also reflect all publically available information about these companies. Strong Form efficiency incorporates all types of information in to the current pricing strategy, which is not yet proved to be present in Indian stock market. For the purpose of statistical analysis of weak form and semi strong form of efficiency in Indian Stock Market the market prices of companies included in the formation of Nifty index was collected from NSE official website. The study was conducted with wide scope both in terms of depth of analysis and breadth of coverage. It has taken a period of 6 years ( ) and daily prices of shares included in the formation of Nifty index. In order to bring more validity to the result, the period in which Indian markets were severely affected by global financial crisis was studied separately. The period under study was 2007 October to 2008 April.

4 Statistical tools like autocorrelation and run test were used to test the weak form market efficiency. One-sample Kolmogorov-Smirnov test was used to find out how well a data series fits a particular distribution. 87 Semi-strong form market efficiency was tested by taking daily returns of companies included in the formation of Nifty Index and compared with the daily Index returns. Beta value for the stocks was calculated to arrive at the residual return. Residual return is the difference between the actual return and expected return. If the difference between the actual return and expected return is zero or near to zero the market is said to be efficient. The formula for calculating expected return was: Expected Return = R i = α i + β i R m + e i, where R m is market index return. The entire study period was divided in to different segments of three months each and the process was repeated for a better result. 5.2 Market Efficiency in the Weak Form Weak form efficiency states that current prices of stocks already reflect all the information that is contained in the historical sequence of prices. Hence there is no benefit in examining the historical prices as far as forecasting the future is concerned. Weak form of market efficiency is popularly called as random-walk theory. If Indian Stock Market is efficient in its Weak form then it is a direct repudiation of technical analysis. Technical analysis relies a lot on historical prices for their future price prediction

5 Weak form efficiency of Indian market during the time frame of 6 years ( ) had been tested using statistical tools like Autocorrelation, and Run test. Daily prices of shares were taken for the study. One-Sample Kolmogorov- Smirnov Test was also used to find out how well a data series fits a particular distribution. 88 Population consisted of all companies listed in NSE. Sample size was 50 companies forming NSE Nifty Index. While doing the pilot study the researcher found that due to constant revisions by NSE, to make the shares chosen for index construction representative of the population, data for only 29 shares were present through out the study period of 6 years. So the Weak form efficiency is studied in two ways; one taking only 29 shares whose data was present through out the study period of six years and the second is taking NSE Nifty index shares for a six year period. 5.3 Test Results of Weak Form of Market Efficiency Study of 29 companies for a period of 5 years on the basis of daily returns The summary statistics of the returns for all the companies included in the study are given in Table 5.1. The normality of distribution is one among the basic assumptions of Weak-form efficient market hypothesis. Mean stock returns are positive with majority of them having comparatively larger volatility (standard deviation).

6 89 Table 5.1 Descriptive Statistics For 29 companies Company N Mean Median Minimum Maximum Std. Deviation ABB ACC BHEL CIPLA GAIL GRASIM HCL HDFC BANK HERO HONDA HDFC ITC ICICI BANK INFOSYS JINDAL M & M MARUTI ONGL PNB RANBAXY RELCAPITEL RELIANCE SIEMENS SBIN SAIL SUN PHARMA TATA MOTORS TATA POWER UNITECH WIPRO

7 Where data are in nominal or ordinal form, or where assumptions about the distribution of data on which a parametric test is based cannot be justified, then non-parametric or otherwise called as distribution-free methods can be used. But parametric tests are more rigorous than non-parametric tests. So to confirm the distributional pattern of the returns, researcher has used Kolmogrov-Smirnov goodness of fit test. 90 Kolmogorov-Smirnov Goodness-of-Fit Test tests whether or not a given distribution is not significantly different from one hypothesised on the basis of the assumption of a normal distribution. This test finds out how well a data series fits a particular distribution. Test compares the cumulative distributional function of the returns with a normal distribution to determine if they are identical. Table 5.2 presents the results of the Kolmogorov-Smirnov Test.It compares an observed cumulative distribution function to a theoretical (Normal) cumulative distribution. Low significance values (<.05) indicate that the observed distribution does not corresponds to the Normal distribution. This confirms that the distribution of Closing Prices is not normal. High significance values (>.05) indicate that the observed distribution corresponds to the Normal distribution and so the distribution of Closing Prices is normal.

8 91 Table 5.2 One-Sample Kolmogorov-Smirnov Test for 29 companies Company Absolute Positive Negative K-S Z p-value ABB ACC BHEL CIPLA GAIL GRASIM HCL HDFC BANK HERO HONDA HDFC ITC ICICI BANK INFOSYS JINDAL M & M MARUTI ONGL PNB RANBAXY RELCAPITEL RELIANCE SIEMENS SBIN SAIL SUN PHARMA TATA MOTORS TATA POWER UNITECH WIPRO

9 Low significance values (<.05) indicate that the observed distribution does not corresponds to the Normal distribution. Thus, the distribution of closing prices is not normal. Majority of the values have low significance values Non Parametric Test Run test for 29 companies The run test can be used to examine the serial independence in share return movements. This test has the advantage of ignoring the distribution of the data, and does not require normality or constant variance of the data. A run can be defined as a sequence of return changes of the same sign. e.g ++ /-- / 0 / -- / has 4 runs. A lower than expected number of runs indicates a market s overreaction to information, subsequently reversed, while a higher number of runs reflect a lagged response to information. Poshokwale, (1996).An abnormally high or low number of runs indicate evidence against the null hypothesis of a random walk.

10 93 Table 5.3 Run Test Result for 29 companies Company Test Value Runs Z-value p-value ABB ACC BHEL CIPLA GAIL GRASIM HCL HDFC BANK HERO HONDA HDFC ITC ICICI BANK INFOSYS JINDAL M & M MARUTI ONGL PNB RANBAXY RELCAPITEL RELIANCE SIEMENS SBIN SAIL SUN PHARMA TATA MOTORS TATA POWER UNITECH WIPRO

11 Here (Table 5.3) the p-values of all the companies are less than So, the null hypothesis that the price movement is not affected by the past price is rejected at 5 percent. The significant negative Z values indicate non-randomness of the series. The result shows that the price movements are not random in behaviour. We can use the historical data for predicting the future prices. The situation suggests that an opportunity to make excess returns exist in the Indian Stock market Parametric Test Auto Correlation Test for 29 companies Researcher employed parametric test i.e. autocorrelation test to confirm the findings of the non-parametric test and to measure the degree of dependency of the series in the Weak form of efficiency during the period under study. Autocorrelation can be defined as the cross correlation of a signal with itself. It is the similarity between observations as a function of the time separation between them. It is a mathematical tool for finding repeating patterns. This method is very often used in signal processing for analysing functions or series of values. Autocorrelation tests show whether the serial correlation coefficients are significantly different from zero. In an efficient market, the null hypothesis of zero autocorrelation will prevail. In this study researcher had tested the correlation between the share price of any period t and t +4,between t and t + 9 and between t and t + 14.To analyse the results,the three limits of correlation of coefficient have been taken. These are ±0 to ±0.25 is low correlation, ±0.25 to ±0.75, moderate correlation and ±0.75 to ±1 is considered to be highly correlated.

12 95 Table 5.4 Autocorrelation Result for 29 companies Company T+4 T+10 T +14 ABB (H) (H) (H) ACC (H) (H) (H) BHEL (H) (H) (H) CIPLA (H) (H) (H) GAIL (H) (H) (H) GRASIM (H) (H) (H) HCL (H) (H) (H) HDFC BANK (H) (H) (H) HERO HONDA (H) (H) (H) HDFC (H) (H) (H) ITC (H) (H) (H) ICICI BANK (H) (H) (H) INFOSYS (H) (H) (H) JINDAL (H) (H) (H) M & M (H) (H) (H) MARUTI (H) (H) (H) ONGL (H) (H) (H) PNB (H) (H) (H) RANBAXY (H) (H) (H) RELCAPITEL (H) (H) (H) RELIANCE (H) (H) (H) SIEMENS (H) (H) (H) SBIN (H) (H) (H) SAIL (H) (H) (H) SUN PHARMA (H) (H) (H) TATA MOTORS (H) (H) (H) TATA POWER (H) (H) (H) UNITECH (H) (H) (M) WIPRO (H) (H) (H) H Highly correlated (±0.75 to ±1) M Moderate Correlation (±0.25 to ±0.75) L Low Correlation (±0 to ±0.25)

13 Table 5.4 shows the autocorrelation coefficients computed for the log of the return series at different lags. Autocorrelation between the prices of shares has been tested for five days, ten days and fifteen days.from the results it is very clear that there is significant autocorrelation at 5 percent significance level among the 29 companies analysed. Results also show that the level of significance decreases by the increase in days compared. 96 For example if we take the autocorrelation value of UNITEC, the value which was and was highly correlated at five day lag decreased to.845 when 10 days lag was considered. The same company s auto-correlation value came to.732 when it came to 5 days lag. The presence of autocorrelation coefficients in the market returns series suggest that there is relationship between past returns and present returns and Indian stock market movements are predictable during the period under study based on past information. 5.4 Study of companies involved in the construction of NSE Nifty Index for a period of 5 years on the basis of daily returns Since the companies involved in the construction of Nifty Index were constantly revised the study had taken two different sample set for finding out market efficiency of Indian stock market.one which included all those companies which were there in the Index construction for the entire study period i.e. six years the results of which is given in the above pages. The second category was companies involved in the construction of index were considered in total. The reason for this comparison was many companies which were there in the index construction did not have their presence for the entire study period. For example AMBUJA was there only from July 2007; similarly RPOWER was there

14 only from 2008 February. Here the data was considered by the researcher in the form of number of observations. 97 The summary statistics of the returns for all the companies included in the study which were there for the entire study period are given in Table 5.5.As mentioned above normality of distribution is one among the basic assumptions of weak-form efficient market hypothesis. Mean stock returns are positive with majority of them having comparatively larger volatility (standard deviation). Table 5.5 Descriptive Statistics for Nifty shares Company N Mean Median Minimu m Maximu m Std. Deviation ABB ACC AMBUJA AXIS BANK BHEL BHARTI CAIRN CIPLA DLF GAIL GRASIM HCL HDFC BANK HERO HONDA HINDALCO HINDUNILVA HDFC ITC

15 98 ICICI BANK IDEA INFOSYS IDFC JP ASSO JINDAL LT M & M MARUTI NTPC ONGL POWER GRID PNB RANBAXY RELCAPITEL RCOM RELIANCE RELINFRA RPOWER SIEMENS SBIN SAIL STER SUN PHARMA SUZLON TCS TATA MOTORS TATA POWER TATA STEEL UNITECH WIPRO

16 Here also to confirm the distributional patterns of the returns, the researcher used Kolmogrov-Smirnov goodness of fit test. The test compared the cumulative distributional function of the returns with a normal distribution to find out whether they are identical. The Kolmogorov-Smirnov Test compared the observed cumulative distribution function to a theoretical (Normal) cumulative distribution. Low significance value (<.05) indicated that the observed distribution does not corresponds to the Normal distribution. Thus, the distribution of closing prices is not normal. High significance values (>.05) indicate that the observed distribution corresponds to the Normal distribution. Thus, the distribution of closing prices is normal. Test result shows that the distribution does not come in the category of normal distribution as the p-values are all less than.05 at 5 percent significance level. 99 Table 5.6 One-Sample Kolmogorov-Smirnov Test for Nifty Shares Company Absolute Positive Negative K-S Z p-value ABB ACC AMBUJA AXIS BANK BHEL BHARTI CAIRN CIPLA DLF GAIL GRASIM HCL HDFC BANK

17 100 HERO HONDA HINDALCO HINDUNILVA HDFC ITC ICICI BANK IDEA INFOSYS IDFC JP ASSO JINDAL LT M & M MARUTI NTPC ONGL POWER GRID PNB RANBAXY RELCAPITEL RCOM RELIANCE RELINFRA RPOWER SIEMENS SBIN SAIL STER SUN PHARMA SUZLON TCS TATA MOTORS TATA POWER TATA STEEL UNITECH WIPRO

18 5.4.1 Non Parametric Test Run test for companies involved in the construction of Nifty The run test was used to examine the serial independence in share return movements. This test has the advantage of ignoring the distribution of the data, and does not require normality or constant variance of the data. In this test the actual number of runs observed in a series of stock price movements is compared with the number of runs in a randomly generated number series. If there is no significant difference between these two, then the security price changes are considered to be at random. 101 From the table (Table5.7) it can be observed that the p-values of all the companies are less than 0.05.So, the null hypothesis that the price movement is not effected by the past price is rejected at 5 percent significant level. The significant negative Z values indicate non-randomness of the series. These results were similar with the test results of 29 companies taken and showed that the price movements were not having randomness in behaviour. So it can be inferred that an investor can make use of historical data for predicting the future prices. Opportunity to make excess returns exist in the Indian stock market Many previous studies on market efficiency have employed run tests in a similar framework such as the studies by Fama (1965), Sharma and Kennedy (1977), Cooper (1982), Chiat and Finn (1983), Wong and Kwong (1984), Yalawar (1988), Ko and Lee (1991), Butler and Malaikah (1992), and Thomas (1995). These studies typically find that in most markets except in Hong Kong, India, Kuwait and Saudi Arabia, the null hypothesis is not rejected.

19 102 Table 5.7 Run Test Results for Nifty shares Company Test Value Runs Z-value p-value ABB ACC AMBUJA AXIS BANK BHEL BHARTI CAIRN CIPLA DLF GAIL GRASIM HCL HDFC BANK HERO HONDA HINDALCO HINDUNILVA HDFC ITC ICICI BANK IDEA INFOSYS IDFC JP ASSO JINDAL LT M & M MARUTI NTPC ONGL POWER GRID

20 103 PNB RANBAXY RELCAPITEL RCOM RELIANCE RELINFRA RPOWER SIEMENS SBIN SAIL STER SUN PHARMA SUZLON TCS TATA MOTORS TATA POWER TATA STEEL UNITECH WIPRO Non Parametric Test Auto Correlation Test for companies involved in the construction of Nifty Numerous studies on market efficiency have reported serial correlation or autocorrelation as one of the significant tool for investigating randomness on stock prices and stock indices. Fama (1965) investigates the behavior of the daily closing prices of the 30 Dow Jones Industrials and finds that the first-order autocorrelation of daily returns are positive for 23 of the 30 firms, which suggests a positive relationship between successive daily returns. Typical recent literature on serial correlation or autocorrelation in return movements includes LeBaron (1992), Sentana and Wadhwani (1992), and Campbell, Grossman, and Wang (1993).

21 104 Table 5.8 Autocorrelation Results for Nifty Shares Company T + 4 T+9 T+14 ABB ACC AMBUJA AXIS BANK BHEL BHARTI CAIRN CIPLA DLF GAIL GRASIM HCL HDFC BANK HERO HONDA HINDALCO HINDUNILVA HDFC ITC ICICI BANK IDEA INFOSYS IDFC JP ASSO JINDAL LT M & M MARUTI NTPC (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (M) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H) (H)

22 105 ONGL (H) (H) (H) POWER GRID (H) (H) (H) PNB (H) (H) (H) RANBAXY (H) (H) (H) RELCAPITEL (H) (H) (H) RCOM (H) (H) (H) RELIANCE (H) (H) (H) RELINFRA (H) (H) (H) RPOWER (H) (H) (M) SIEMENS (H) (H) (H) SBIN (H) (H) (H) SAIL (H) (H) (H) STER (H) (H) (M) SUN PHARMA (H) (H) (H) SUZLON (H) (H) (H) TCS (H) (H) (H) TATA MOTORS (H) (H) (H) TATA POWER (H) (H) (H) TATA STEEL (H) (H) (H) UNITECH (H) (H) (M) WIPRO (H) (H) (H) H High correlation (±0.75 to ±1) M Moderate Correlation (±0.25 to ±0.75) L Low Correlation (±0 to ±0.25) The results of the tests indicate the existence of high correlation between the share prices for majority of the companies included in the study. The estimated serial correlation or autocorrelation is presented in Table 5.8

23 The study which is presented in this chapter seeks evidence supporting the existence of weak-form efficiency of Indian market. The sample included the daily closing price of all the shares included in the formation of Nifty Index. The study period was from The null hypothesis of the study was whether the Indian Stock Market is weak form efficient. The results of both nonparametric (Kolmogrov Smirnov goodness of fit test and run test) and parametric test ( Auto-correlation test )provide evidence that the share prices do not follow random walk model and the significant autocorrelation co-efficient at different lags reject the null hypothesis of weak-form efficiency. 106 The results are consistent in different sub-sample observations and for individual securities. The issues are important to security analysts, investors and to security exchange regulatory bodies in their policy making decisions to improve the market condition. This study deserves a continuous research on this area to reach an ultimate conclusion about the level of efficiency of emerging markets like India market. 5.5 Semi-strong Market Efficiency of Indian Stock Market Semi strong market efficiency is part of Efficient Market Hypothesis which implies that all publicly available information is calculated into a stock's current share price. This means that neither fundamental nor technical analysis can be used to achieve superior gains. In an efficient market, when a new piece of information is added to the market, its implications for security returns are instantaneously and un biasedly impounded in the current market price. In other words it can be said that a capital market is efficient if the corporate event announcements like stock split, buyback, right issue, bonus announcement,

24 merges & acquisitions, dividend etc are quickly and correctly reflected in the security s prices. 107 In the second part of this chapter researcher presents the results of the test of Semi-strong efficiency of Indian Stock market. The study had been conducted on 29 companies shares whose data were present through out the study period of 6 years. 5.6 Test Result of Semi-strong efficiency of Indian Stock Market Semi-strong efficiency tests deal with whether or not security prices fully reflect all publically available information. All these tests attempt to experiment whether share prices react quickly and correctly to a new piece of information. If the results give evidence that share prices do not react adequately and quickly to the various information, it means that the market offers opportunities for earning superior returns. An investor can earn excess returns by using this publicly available information. Some of the earlier studies conducted in testing Semi-strong form of market efficiency have been contributed by Fama, Fisher and Jense. Methodology followed in various studies testing Semi-strong market efficiency is to take an economic event and measure its impact on the share price. The impact is measured by taking the difference between actual return and expected return on a security. This is known as the residual analysis. Excess return would be present if there is a positive difference between the actual return and expected return. In the present study also the researcher had used the residual analysis model suggested by William Sharpe.

25 108 The formula used for calculating Expected return (R i ) R i = α i + β i R m + e i Where: R i = Expected Return of the i th stock α i = Intercept β i = Beta value of the i th stock R m = Return of the market index e i = The error factor The formula used for calculating Actual Security return = Today s security return Today s price Yesterday s price *100 Yesterday s Price Today s market return = Today s index Yesterday s index *100 Yesterday s index Systematic risk is the variability in security returns caused by economic or other market factors. All securities traded in the market will be affected by such changes. But some of them exhibit greater variability while others have some minor variations. The securities which are affected to a greater extend are said to have higher systematic risk. Systematic risk is measured by relating the security s variability with the variability in the market index.

26 Beta is the statistical measure of the risk of a security. A security can have positive, negative or zero beta value. Lager the volatility of a share, larger will be the beta value for that share. A beta of 1.0 indicates a security of average risk. If beta value is more than 1.0 it has above average risk. Alpha is the difference between the actual return produced by an investment and the rate that might have been expected, given its level of beta. Beta expresses risk in relation to the market as a whole and its value can be positive or negative, but in practice it tends to fall between and as: The formula used for finding the beta and alpha co-efficient can be expressed β = n X Y ( x) y) Where: n X 2 - ( X ) 2 X = NSE Index Y = Closing price of the security x = Index return y = security return α = Y - β X Residual Return = Actual return Expected Return (Residual return will be positive if the actual return is more than the estimated return)

27 If the excess return or residual return is close to zero, it implies that the price reaction following any of the public announcements is immediate and price adjusts quickly to the new level. If the excess return is zero or near to zero it would validate the presence of Semi-strong form of market efficiency. 110 The following tables give the test result of Semi-strong form of market efficiency. Tests have been conducted using daily closing price of the 29 companies shares whose data was available for the study period of six years. The entire study period was split in to three months each and the process was repeated for better results. Residual mean indicate the mean of the residual returns on a daily basis for the period under study.

28 111 Table 5.9 Test of Semi-strong form of market efficiency for Jan 2004 Mar 2004 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

29 112 Table 5.10 Test of Semi-strong form of market efficiency for Apr 2004 Jun 2004 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

30 113 Table 5.11 Test of Semi-strong form of market efficiency for Jul 2004 Sep 2004 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

31 114 Table 5.12 Test of Semi-strong form of market efficiency for Oct 2004 Dec 2004 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

32 115 Table 5.13 Test of Semi-strong form of market efficiency for Jan 2005 Mar 2005 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

33 116 Table 5.14 Test of Semi-strong form of market efficiency for Apr 2005 Jun 2005 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

34 117 Table 5.15 Test of Semi-strong form of market efficiency for Jul 2005 Sep 2005 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

35 118 Table 5.16 Test of Semi-strong form of market efficiency for Oct 2005 Dec 2005 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

36 119 Table 5.17 Test of Semi-strong form of market efficiency for Jan 2006 Mar 2006 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

37 120 Table 5.17 Test of Semi-strong form of market efficiency for Apr 2006 Jun 2006 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

38 121 Table 5.18 Test of Semi-strong form of market efficiency for Jul 2006 Sep 2006 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

39 122 Table 5.19 Test of Semi-strong form of market efficiency for Oct 2006 Dec 2006 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

40 123 Table 5.20 Test of Semi-strong form of market efficiency for Jan 2007 Mar 2007 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

41 124 Table 5.21 Test of Semi-strong form of market efficiency for Apr 2007 Jun 2007 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

42 125 Table 5.22 Test of Semi-strong form of market efficiency for Jul 2007 Sep 2007 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient

43 126 Table 5.23 Test of Semi-strong form of market efficiency for Oct 2007 Dec 2007 Company Residual Mean N Result ABB Efficient ACC Efficient BHEL Efficient CIPLA Efficient GAIL Efficient GRASIM Efficient HCL Efficient HDFC BANK Efficient HERO HONDA Efficient HDFC Efficient ITC Efficient ICICI BANK Efficient INFOSYS Efficient JINDAL Efficient M & M Efficient MARUTI Efficient ONGL Efficient PNB Efficient RANBAXY Efficient RELCAPITEL Efficient RELIANCE Efficient SIEMENS Efficient SBIN Efficient SAIL Efficient SUN PHARMA Efficient TATA MOTORS Efficient TATA POWER Efficient UNITECH Efficient WIPRO Efficient