6516 Technical

Technical Analysis in the Indian Capital Market …A Survey Sanjay Sehgal* & Meenakshi Gupta**

Introduction The Indian economy has been going through a phase of economic transition since 1991.The Indian capital market has received special attention under the policy of liberalisation. Reforms in the security market, particularly, the establishment and empowerment of Securities and Exchange Board of India (SEBI), abolition of Controller of Capital Issues of India (CCI), market determined allocation of resources, screen based nationwide trading, market determined interest rate structures, entry of Foreign Institutional Investors (FII) and Non Resident Indians’ (NRI) investments, Mutual fund entry in the private sector, dematerialisation and electronic transfer of securities, rolling settlement, sophisticated risk management and derivatives trading have greatly improved the regulatory framework and efficiency of trading and settlement in the Indian capital market. The Indian market is now comparable to many developed markets in terms of a number of quantitative and qualitative parameters. The Indian stock market consists of 23 stock exchanges including two exchanges set up in the reforms era i.e. National Stock Exchange of India (NSE) and Over The Counter Exchange of India (OTCEI). NSE was inaugurated in 1994 with objectives to provide a nationwide trading facility, to meet international

The survey aims at providing insights about the way technical traders operate in the financial market and the trading strategies that they adopt. The survey covered institutional and individual technical traders with a long and active trading record for the Indian market. It is observed that the sample respondents tend to use technical analysis along with fundamental analysis for security selection. They fit the technical tools mainly on the equity segment of the market and relatively prefer their use during the market upturns. They further feel that volume indicators provide independent information compared to price indicators. They seem to be using the classical technical tools more frequently while sophisticated technical trading systems as well as time series econometric tools were relatively less important to them. They believe that the choice of tools is not related to company characteristics such as size, relative distress and leverage.

*Professor in Finance, Department of Financial Studies,University of Delhi, South Campus. e-mail: [email protected] **Lecturer in Commerce, Sri Aurobindo College, University of Delhi. e-mail: [email protected]

Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

92

securities market standards, to shorten settlement cycles and to impart fairness, efficiency, and transparency to the securities trading. OTCEI was set up in 1992 to promote trading in an entirely new set of companies sponsored by the members of the OTCEI. It allows listing of small and medium sized companies (with an issued capital of Rs.3 million to Rs.250 million). An Interconnected Stock Exchange (ISE) has also been created by 14 regional stock exchanges, to provide cost effective trading linkage to all the members of the participating exchanges, with the objective of widening the market for the securities listed on these exchanges. ISE aims to address the needs of small companies and retail investors with the guiding principle of optimising the existing infrastructure and harnessing the potential of regional markets so as to transform these into a liquid and vibrant market through the use of state-of-the-art technology and networking. At present all the 23 stock exchanges in the country offer a screen based trading system. The trading system is connected using the VSAT (Vary Small Aperture Terminals) technology from over 357 cities. It has grown exponentially with an all India market capitalisation of Rs.13187,953 million, 9413 listed companies, 9368 members 20 millions of investors, and a turnover of 122.9% at the end of March 2004 (Indian Securities Market Review, 2004). The market has witnessed fundamental institutional changes resulting in drastic reduction in transaction costs and significant improvement in efficiency, transparency and safety. At the end of 2003, Standard and Poor (S&P) ranked the Indian stock market 25th in terms of market capitalisation, 16th in terms of total value traded in stock exchanges and sixth in terms of turnover ratio (S&P Emerging Stock Market Fact Book, 2004). Except for the USA, India has the highest number of securities listed on stock exchanges. With this expanding size, there is a need for comprehensive analysis of the Indian securities market. The mature markets of the world have been heavily researched in the last few decades. However, similar research activity for emerging markets like India is very limited. The research gap is even more pronounced on the short end of the market that involves understanding investor behaviour in terms of framing short-term investment strategies with the help of a technical trading system. Hence, there is a need to perform a survey of the market practitioners before proceeding to empirically evaluate the technical indicators. The survey aims at determining the attitude of the analysts towards technical analysis and the success they attribute to it. It will also highlight the technical indicators they are using in different market phases and their preferences for using them. The study is organised into six sections, including the present one. The next section gives an overview of technical analysis. It is followed by a review of

Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

93

literature, the data and methodology and the survey findings. The summary and conclusions are given in the final section. An Overview of Technical Analysis There have been two main approaches to analyse the securities market - the fundamental approach and the technical approach. According to the fundamental approach, each security has an intrinsic value and the market forces would ensure that the share price converges to that value in the long run. The technical approach states that past share prices and volumes tend to follow a pattern and they can be used to predict future price movements. Forces of demand and supply in general determine the share price. The fundamentalists, however, think that demand and supply are a function of rational factors, while technicians attribute it to psychological factors. The technical approach to investment is essentially a reflection of the idea that prices move in trends, which are determined by the changing attitudes of the investors towards a variety of economic, monetary, political and psychological forces. The technical approach is based on the theory that the price is a reflection of mass psychology (the crowd) in action. It attempts to forecast future price movements on the assumption that crowd psychology moves between panic, fear, and pessimism, on one hand, and confidence, excessive optimism and greed on the other. But the possible existence of an efficient market1 poses a major challenge to the ability of technical analysts to select shares, which will outperform the market. If technical analysts can produce excess returns they must have identified information not reflected in security prices. If such information is identifiable, market prices do not reflect all available information and hence the market is not efficient. The tools of technical analysis can be divided into two main groups, namely classical technical analysis and modern technical analysis. Classical technical analysis is based on chart patterns and the analyst tries to identify price patterns. Their goal is to profit from trading when patterns occur. Some of the important Chart Patterns are Head and Shoulders, Support and Resistance, Gap Analysis, Trend lines, Triangles, Rectangles, Double Top and Double Bottom. Other classical tools are Elliot Wave Principle, Fibonacci Ratios and Candlesticks Chart Patterns.

1.

Efficient market theory is based on the notion that nobody can outperform the market because any price at any given moment incorporates all available information. 2.

A trend is a condition where prices are rising or falling, on balance, and where there is a definite up or down direction to a price move. Successively higher highs and higher reaction lows define an up trend. Successively lower lows and lower reaction highs define a downtrend. An up trend is also called as bull phase and a downtrend is called a bear phase.

Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

94

Modern technical analysis is based on 3 major groups of indicators that help in identifying trends and their turning points. They can be trend following indicators which work best when market is in a trend2, oscillators which work best when the market is in trading range, and miscellaneous indicators such as aggregate market analysis tools, psychological indicators, trading systems, etc. Trend Following Indicators can be divided into 3 groups i.e. Moving Average (MA), Directional System and Trend Following Volume indicators. Oscillators are used to analyse overbought/oversold conditions, divergences and crossovers. An oscillator becomes overbought when it reaches a high level associated with tops in the past. It means too high, ready to turn down. An oscillator becomes oversold when it reaches a low level associated with bottoms in the past. It means too low, ready to turn up. Horizontal reference lines on the charts mark these levels. Divergences occur when the oscillator and the price action diverge amongst themselves. Crossovers involve the use of two different moving averages that comprise of a slow moving average and a fast moving average. When the fast average crosses above the slower moving average, then, a buy signal is generated. However, when the faster moving average crosses below the slower moving average, then, a sell signal is generated. Some of the important oscillators are Commodity Channel Index (CCI), Relative Strength Index (RSI), Stochastic, Rate of Change (ROC), and William %R. A miscellaneous group includes all other indicators used by practitioners for the aggregate market as well as individual security analysis. Review of Literature The technical analysis (TA) approach to capital market evaluation has received little attention and acceptance as compared to fundamental analysis (FA). But in recent years the popularity of technical school of thought is increasing among academicians and practitioners. A brief review of literature is given below for studies abroad and studies for the Indian market. Studies Abroad Alexander (1961) examined the profitability of using a mechanical filter rule for stock trading. His study indicated substantial profits from filters. Cootner and James Jr (1962) tested the Moving Average technique. They believed that this technique averages out random fluctuations and thus changes in the direction of basic trends can be isolated. Fama and Blume (1966) conducted a study of the thirty Dow Jones Industrial Stocks using 24 different filters, ranging in size from 0.5 percent to 50 percent for the time period from 1957 to Sept 1962. The conclusion of their test was that filter rules were not profitable when the effect of interim dividends and brokerage commissions are considered. Levy (1967) tested the Relative Strength technique. This technique selected those stocks that were performing the best, relative to their Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

95

average price of the previous 26 weeks. According to Levy, the relative strength portfolios had higher returns than B-H strategy, but also the risks were higher. James Jr (1968) extended the moving average tests to include averaging by exponential smoothing. To test the significance of his findings, the difference Di, between the moving average return and the buy and hold return was calculated for each sample security. The results show that none of the rules or various smoothing coefficients was more successful than the B-H strategy unless the data is unadjusted for dividends. Neftci and Policano (1984) used the moving average and slope method to analyse the future market. The results suggest that by characterising the actual behaviour of the market participants, improved price predictions can be obtained in future markets. Dawson (1985) tested whether investors could have outperformed the market by using actual share recommendation based solely upon technical analysis made by a specific investment advisory firm. Share recommended by technical analysis outperformed the market but after adjusting for trading commissions, market trends and risk, the recommended share did not outperform the market. Brock, Lakonishok and Le Baron (1992) tested the two simplest and most popular trading rules- Moving Average and Trading Range Break (Resistance and support level) – by utilising the Dow Jones Index from 1897 to 1986. The results provide strong support for technical strategies. Consistently, buy (sell) signals generate returns which are higher (lower) than normal returns. Blume, Easley and O’Hara (1994) investigated the informational role of volume and its applicability for technical analysis. In their model, technical analysis is valuable because current market statistics may be sufficient to reveal some information, but not all. The uncertainty in the economy is not resolved in one period and sequences of market statistics can provide information that is not impounded in a single market price. Unique to their model is the feature that volume captures the important information contained in the quality of traders’ information signal. Batten and Ellis (1996) analysed the technical trading performance and weak form efficiency of the Australian All Ordinary Index (AAOI). The trading systems employed were able to generate a return greater than B-H strategy without considering transaction costs. Sullivan, Timmermann, and White (1999) used bootstrap methodology to evaluate simple technical trading rules while quantifying. the data snooping bias and fully adjusting for its effect in the context of the full universe from which the trading rules were drawn. Evaluation is based on mean returns and on the Sharpe ratio, which adjusts for total risk. The best technical trading rules for the mean returns from the full universe are from the short-term price movements (5 day moving average, 2 day on-balance-volume). The best trading rule by the Sharpe ratio criterion is also

Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

96

based on relatively short sample using 2 to 20 days of price information. In the full universe the best performing trading rule earned a mean annualised return of 17.17 percent resulting from 6310 trades. Mamaysky and Wang (2000) proposed a systematic and automatic approach to technical pattern recognition using non-parametric kernel regression, and applied this method to a large number of US stocks. They found that over a 31-year sample period, several technical indicators do provide incremental information and may have some practical value. Studies of the Indian Capital Market Sehgal and Garhyan (2002) evaluate whether share recommendations based on technical analysis provide abnormal returns in the Indian capital market. Several return measures have been employed including those adjusted for market trend, risk and transaction costs. The study involves 21645 recommendations for 21 companies using 13 technical indicators. The mean return was found statistically significant for the total period. But the gains disappear in the case of market-adjusted measures. The returns were found significant for the risk-adjusted measures and also after the adjustment for transaction costs. Mitra (2002) examined the applicability of moving average based techniques and filter rule technique for investments on trading in Indian stock market. The study found that profit is high in moving average crossover with periods of 2 and 10 days. All low value filters give profits; however, the profitability becomes negative with high value filters. The filter of 1 percent of stock price was found suitable for all the series and the use of filter method was found less risky. These studies attempt to obtain unusual profits by using technical tools. Several tools have shown the ability to provide a higher return than that of buy and hold policy. But after adjusting for trading commissions or transaction costs, the results are not so encouraging for a developed market; however, Indian evidence on technical trading system seems to be more supportive. The disparity in empirical findings for developed markets and the Indian market may probably reflect differences in the levels of efficiency of these markets. Data and Methodology The survey aims at determining the attitude of the analysts towards technical analysis and the success they attribute to it. It will also highlight the technical indicators they are using in different market phases and their preferences for using them. The sample consists of 25 respondents. The respondents are investment analysts, mutual fund managers, brokers and active technical traders. It is a representative set of active investors, as in our study, we are looking for a specialised sample involving respondents Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

97

with a long experience in the stock market and a reasonable length of time spent on technical trading. The average experience of respondents in the stock market is 9 years and in technical analysis 7 years, as shown in figure 1. The questionnaire was distributed personally by hand or through e-mail to 50 individuals. Of those approached, fifty percent returned it after filling in the necessary details. A standard questionnaire (see Appendix) was framed which contained mostly closed

Figure 1 ended questions. However, some open-ended questions were also incorporated wherever it was necessary to obtain a detailed view of the respondents with regard to a given issue. The aim was to accumulate views and opinions of the respondents in order to gain deeper insight into the working of the stock market, in general, and the economic feasibility of technical trading systems, in particular. The questionnaire was sent to three experts for validation check and their suggestions were duly taken into account before furnishing it to the respondents. The questionnaire was framed in three parts. Part one consisted of the personal profile of the respondents. Part two dealt with questions regarding the use of technical analysis tools on different securities and part three laid emphasis on the market index used, and also if the analysts made any distinction regarding different sectors of the industry and different company characteristics. A set of fourteen questions were asked in part two and only six in part three. We used convenient but judgemental sampling as the desired persons were selected on the basis of personal contacts. Survey Findings Technical analysis – Individual Securities All the respondents have faith in technical analysis and agree that it helps in generating better returns. Almost two thirds of the respondents are using fundamental analysis along with technical analysis as shown in Figure 2. It is probable that they are using FA for selection of securities based on their intrinsic value and TA for market timing purposes. Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

98

Figure 2 All twenty-five respondents are using one or more technical software packages at one time to analyse the securities. Their cumulative responses are given below in Table1. It is evident that metastock is the most popular software package among the analysts. It provides a variety of tools for decision-making purposes. But almost all of them are using one or the other software along with metastock.

Table1 List of Software Used Software ASA Advanced Get Money IRIS Metastock Aspen Graphics Omega Elwave Wave Analyser

Decision, Vol. 32, No.1, January - June, 2005

No. of users 4 4 2 5 23 2 5 5 3

Technical Analysis in the Indian Capital Market …A survey

99

All the respondents are using technical indicators in all the three phases i.e. Bull, Bear and Normal (also known as trading range). The Bull Phase is symbolised by a prolonged rise in the prices of shares sustained by the buying pressure of investors. The Bear Phase is symbolised by a prolonged period of falling prices, dominated by selling pressure in the market place.The Normal Phase is a period where prices move sideways or in a lateral direction or where a series of highs peak in the same area and a series of lows bottom around the same levels. It appears from the survey that technical indicators are being used more extensively in the Bull Phase as shown in Figure 3.

Figure 3 In the Bull Phase about 88% are using technical tools extensively and 12% are using them moderately. In the Bear Phase, 64% are using them extensively, 32% moderately and only 4% are using them less extensively. In the Normal Phase, about 56% are using them extensively, 32% moderately, and only 12% are using them less extensively. Investors suffer losses in the Bear Phase so their focus shifts more to understanding and analysing corporate fundamentals. As short selling is not allowed they cannot use technical tools in the Bear Phase except for the stocks on which stock futures are available. In the Normal Phase, the market is in trading range and profit opportunities are less lucrative and it is always advisable to use oscillators. The less extensive use of TA in the Normal Phase is also implied by the limited use of oscillators by the Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

100

respondents, which we observe in a later question. All the respondents are using technical tools on shares. 50% of them are using their tools on shares and financial derivatives (FD), and only 16% are using them on shares, FD, forex and bonds. Thus it seems that TA is predominantly used for the equity sector of financial markets, while in other sectors, its use is limited. Sixty percent of the respondents feel that the settlement system does not influence the technical recommendations while the remaining forty percent feel otherwise, as shown in Figure 4.

Figure 4 The reason they provide against the relationship between settlement system and technical recommendations is that TA is based on past trends and hence it has nothing to do with settlement period. Also, as TA is based on demand and supply of the instrument, factors like end of settlement /long trading weekends prompt liquidation of positions by short term traders. As active stocks are on rolling settlement system, the impact of the weekly settlement system is not there. In the rolling settlement system, the pressure is not on a particular day as against the weekly settlement system where pressure was on a particular day (settlement day). In an earlier study by Sehgal and Garhyan (2002), the average holding period was found to be 5.4 days, which was approximately equal to the length of the weekly settlement period. The rolling settlement system may not affect the holding period as investors can settle their accounts on a continuous basis. About 36% of respondents say that circuit breakers and circuit filters affect the return and 64% of them say it does not affect the return, as circuit breakers are Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

101

relevant for intraday or index securities as shown in Figure 5.

Figure 5 The respondents are trading in BSE SENSEX and NIFTY securities and filters are not applicable on them and circuit breakers are applied to the index as a whole, which are applicable infrequently i.e., only in the event of a major market breakout. About 40% of the respondents are using both arithmetic and log scale3 and 28% are using only log scale and 32% are using only arithmetic scale as given in figure 6. Use of arithmetic scale means the linear scale but sometimes when values on one axis grow at an exceptional rate, use of log scale is preferred.

Figure 6 3

Logarithmic scaling is the type of price scale where equal percentage moves are measured equally. For example, a move from 10 to 20, representing a gain of 100 percent, will be 5 times as long on the logarithmic price scale as a move from 50 to 60, a gain of 20 percent. In arithmetic scale both these price moves would be equal on the price scale.

Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

102

The respondents made no distinction in using the tools in different market phases.They are using all the tools in all the phases. The only distinction they made is their preferences for tools as given in the Table 2 below. Table 2 Tools used with high preference in all 3 phases 1

Trend lines

2

Gap Analysis

3

Japanese Candlesticks

4

Fibonacci numbers

5

Moving average

6

RSI

7

CCI Tools used with medium preference in all 3 phases

1

Elliot Waves

2

MACD

3

Open Interest Tools used with low preference in all the 3 phases

1

Directional indicator

2

Bollinger Band

3

MACD-Histogram

4

ROC

Preference is shown towards classical indicators and then towards the trend following indicators. Despite so much of emphasis on modern technical indicators in academics, such as directional system, stochastic, momentum, William %R, elder-ray etc. the technicians seem to prefer classical tools of technical analysis. This could be owing to the fact that technicians do not have enough faith in modern technical analysis. The second reason may be that they have already evaluated the economic profitability of modern indicators vis-à-vis classical indicators and have found the later to be more useful. But, if it is owing to the first reason, then modern technical tools need to be evaluated. Unlike so much of emphasis in theory about the Triple Screen Trading Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

103

System, respondents are not using them. However, some of them are using channeltrading systems like Commodity Channel Index (CCI) and Bollinger Band systems. 80% of the respondents are using both price and volume indicators but only 4% of the respondents are using only volume indicators and 16% are using only price indicators as shown in Figure 7.

Figure 7 From the responses, it is clear that although the respondents are using both the volume indicators and price indicators, the latter is preferred to the former as shown in Figure 8.

Figure 8 Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

104

68% of the respondents exhibit no preference for volume indicators over price indicators and 56% believe that volume indicators do not lead price indicators. 60% of the respondents are of the opinion that volume indicators and price indicators give different signals and 72% of them say that volume indicators do not confirm (give similar results) price indicators. It seems that price and volume indicators are not related. They give independent information to investors. There seems to be no leadlag relationship between them. Only 8% of the respondents are using time series models, such as, ARIMA (Auto Regressive Integrated Moving Average) and VAR (Vector Auto Regression) along with technical analysis as given in Figure 9.

Figure 9 Econometricians regularly use and recommend time series models like ARIMA and VAR for forecasting financial time series like stock prices, interest rates and forex movements. Practitioners tend to rely on technical trading tools for forecasting, as the time series models cannot explain the returns generated by these tools (Brock, Lakoniskok, and LeBaron,1992). It may be possible that they are not using time series models at all or have abandoned their use after trying them for a considerable time. The probable likelihood is that the first reason is more valid. A comparative evaluation should be done of the economic feasibility of technical trading systems and time series models. This will hopefully fill the gap between theory and practice. Table 3 below shows the shares on which the respondents strongly feel that technical analysis works. The holding period of the shares varies from 3 to 15 days. Of these, 16 shares are from the BSE Sensex as on Dec 2003 and all of them are Nifty stocks.

Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

105

Table 3 Name of the Shares that Technicians are Using S.no. 1 2 3 4 5 6 7 8 9 10

Name ACC BPCL Grasim HCL Tech Hindalco HLL HPCL Infosys ITC Hughes

S.no. 11 12 13 14 15 16 17 18

Name M&M OBC ONGC Reliance Satyam Computers State Bank Zee Telefilms Tata power 19 TELCO 20 TISCO

Table 4 gives various financial parameters such as size (market capitalisation which is measured as market price times the number of shares outstanding), trading volume, Market Price to Book Equity Ratio (P/B), Price Earning Ratio (P/E), and Debt Equity Ratio (D/E) for these twenty shares. These parameters are calculated as on the 31st of December 2003 because the survey has been conducted during this time period. The ratios are calculated by dividing each value of the recommended company with the average value of the BSE500 Index. To get an idea of how high or low these ratios are for a given company, these ratios are converted into indices. These are big capitalisation companies with an average daily high trading volume and low Debt-Equity ratio (D/E), low Price to Book Value (P/B) and low Price to Earnings ratio (P/E). It means that investors are doing value investing in big size companies with low risk. Probably they are ignoring strong profit opportunities that may arise from choosing value bargains in the small and mid capitalisation sector of the market.

Decision, Vol. 32, No.1, January - June, 2005

106

Technical Analysis in the Indian Capital Market …A survey

Table 4 Financial Parameters for the Selected Companies Co. Name Size Trdg.Vol. ACC 0.08 0.03 BPCL 0.13 0.22 Grasim 0.83 0.94 HCL Tech 5.05 1.37 Hindalco 0.11 0.06 HLL 0.13 0.38 HPCL 0.08 0.02 Infosys 0.51 0.07 ITC 0.05 0.22 Hughes 7.08 4.01 M&M 1.45 2.31 OBC 1.85 3.34 ONGC 46.57 3.80 Reliance 33.29 28.25 Satyam Comp 4.86 37.72 State Bank 13.18 18.86 Zee Telefilms 2.82 6.82 Tata power 2.18 3.10 TELCO 5.18 11.44 TISCO 5.39 22.72 Average 6.54 7.28

P/B 0.22 0.24 0.66 1.29 0.54 0.51 0.34 0.92 0.47 0.73 0.83 0.63 0.86 0.82 1.48 0.46 0.48 0.42 1.49 1.31 0.74

P/E 0.20 0.12 0.18 0.51 0.18 0.27 0.33 0.57 0.13 0.28 0.42 0.20 0.26 0.38 0.71 0.18 1.37 0.24 0.53 0.24 0.36

D/E 0.66 0.31 0.32 0.00 0.18 0.00 0.09 0.00 0.01 0.00 0.34 0.25 0.01 0.33 0.00 0.35 0.05 0.25 0.26 0.41 0.19

It is interesting to note that when we find out the returns of the top 30 and bottom 30 shares of the BSE500 index on the basis of company characteristics such as size, P/ B and P/E separately, an alarming difference in returns is seen. We restricted our empirical evaluation to a subset of recommended indicators. The selected indicators are inbuilt in metastock (a popular technical software) and hence easy to compute and analyse. The list of the technical indicators used is given in Table 5.

Decision, Vol. 32, No.1, January - June, 2005

107

Technical Analysis in the Indian Capital Market …A survey

Table 5 List of technical indicators used Indicator Used Moving average Directional Indicator MA Convergence Divergence Stochastic Relative Strength Index Bollinger Band Commodity Channel Index Negative Volume Index

Symbol MA DI MACD STOCH RSI BB CCI NVI

The returns for the year 2003 for these technical indicators and for simple Buy and Hold (SBH) strategy are shown in Table 6. The annual % return on individual stocks for an indicator is calculated as: N Σ [(Pt-Pt-1)/Pt-1]*100, ………(5.1) t=1 where N is number of trades. Pt =Price at the end of each trade Pt-1=Price at the beginning of each trade The annual return on the top 30 and bottom 30 portfolios are then computed by taking a simple average of the returns on individual stocks belonging to the concerned groups.

Table 6 Profits (in %) for the Year 2003 Indicator MA DI MACD STOCH RSI BB CCI NVI SBH

Small Size 245.98 261.37 217.08 348.71 47.26 20.64 209.50 61.19 333.97

Big Size 50.60 63.63 28.60 98.57 21.53 5.10 18.64 8.07 84.42

Low P/B 188.63 159.93 177.43 215.31 17.61 22.08 43.94 45.94 220.42

Decision, Vol. 32, No.1, January - June, 2005

High P/B 6.12 2.56 13.31 17.55 4.68 12.36 16.05 -6.05 20.77

Low P/E 157.13 170.11 150.07 226.71 42.28 26.68 78.64 17.64 300.50

High P/E 39.93 30.36 53.68 56.75 0.44 12.77 11.42 9.48 69.71

Technical Analysis in the Indian Capital Market …A survey

108

The returns on small stocks were almost 5 times higher than that of big stocks. The higher return on small stocks may be due to higher risk exposure because of operating financial and liquidity risk. But, still the difference in returns may not be justified by the risk alone and ignoring this difference might be a mistake. Similarly, for low P/B shares, the returns were 10 times higher than that of high P/B shares and for low P/ E shares, the returns were 4 times higher than that of high P/E shares. However, the magic wanes out when we compare the returns provided by different technical indicators with those on a SBH strategy as shown in Table 6. The SBH strategy seems to outperform almost all technical indicators for all the portfolios.

Market, Sectoral and Company characteristics related patterns. In this section, we attempt to find out how technical analysts perform aggregate market analysis, and select technical indicators for specific industrial sectors and companies with different characteristics. Almost all of them are using more than one index as a barometer of the market. Fifteen out of twenty-five respondents are using BSE500 and twelve respondents are using BSE Sensex. Nine of them are using NSE, 10 of them S&P CNX Nifty and only two are using BSE100 and none of them is using BSE200. Respondents’ cumulative responses are given below in Table 7.

Table 7 Indices Used by the Respondents SENSEX BSE100 BSE200 BSE500 S&PCNX Nifty NSE

Decision, Vol. 32, No.1, January - June, 2005

12 2 0 15 10 9

Technical Analysis in the Indian Capital Market …A survey

109

As Sensex and Nifty are price sensitive indices and BSE500 is supposed to be a surrogate for aggregate economic wealth, they are using a combination of these two as a market barometer. 92% of the respondents are using the same set of tools to test the market index as well as securities and only 8% are using a different set of tools as given in Figure 10.

Figure 10. The respondents use New High-New Low (NH-NL) and advance decline (A/D) indicators for forecasting the direction of the market. However, since an index itself is a basket of securities, the respondents prefer using the same set of technical tools, which they apply for individual stocks, for evaluating the market. Table 8 gives the list of industries, which the respondents prefer to use. The companies recommended by respondents as shown in Table 3, also fall within these recommended sectors. They are trading in both old economy and new economy shares.

Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

110

Table 8 Preferred Industrial Sectors S.No.

Industry

1

Banking

2

Capital Goods

3

Cement

4

Entertainment

5

FMCG

6

IT

7

Power

8

Refineries

9

Steel

10

Telecom

Respondents use all indicators for all categories of stocks. No distinction is made in using different tools for different categories of stocks based on company characteristics. This could be owing to the fact that investors are applying TA on specific sets of stocks (as shown in Table 3) i.e., stocks of big companies with high trading volume, low leverage and valuation ratios. Hence, they are not able to evaluate the sensitivity of technical indicators across the stocks exhibiting a different set of company characteristics. To study the effect of company characteristics on returns of stocks, we classified the BSE500 stocks into 3 equal groups on the basis of size i.e.S1 (small), S2 (medium) and S3 (big). We further classified the same stocks into 3 equal groups on the basis of their P/B ratios i.e. V1 (low), V2 (medium) and V3 (high). We formed double sorted equally weighted portfolios based on size and value groupings i.e. S1V1, S2V1, S3V1, S1V2, S2V2, S3V2, S1V3, S2V3, and S3V3, where S1V1 is a portfolio of small stocks with low valuation ratios and S 3V3 is a portfolio of big stocks with high valuation ratios. The equally weighted portfolios are more desirable as their parameters are loaded with less measurement errors (Lakonishok, Shleifer, Vishny 1994). We estimate the returns provided by each of these portfolios using 8 different technical indicators for the calendar years 2002 and 2003 as shown in Tables 9 and 10 respectively. The trading system in India has changed w.e.f. July 2001 where in the badla system was abolished. Hence, evaluation for the previous period may not be useful owing to comparison problems. Returns shown in Tables 9 and 10 are net of transaction costs Decision, Vol. 32, No.1, January - June, 2005

111

Technical Analysis in the Indian Capital Market …A survey

i.e., the commission charged by security brokers and other sales agents on buying and selling the securities. Daily price data (using arithmetic scale) is used for calculating the portfolio returns. The unadjusted returns (shown in Panel A) on portfolios are calculated in two steps. First, the annual % return on individual stocks is calculated using the equation 5.1 and in the second step, annual portfolio returns are estimated by taking a simple average of security returns. Table 9 Profit (in %) for the year 2002 for different size and value portfolios Panel A Unadjusted Returns MA

DI

MACD

STOCH

RSI

BB

CCI

NVI

SBH

S1V1

72.91

117.56

62.20

135.07

45.14

9.73

39.70

57.66

174.75

S2V1

98.62

142.05

65.88

110.30

34.24

-1.39

53.72

28.16

135.13

S3V1

70.23

125.84

47.45

66.50

24.93

8.27

19.58

26.51

124.69

S1V2

52.25

82.39

35.25

56.13

11.06

7.01

12.22

11.72

154.45

S2V2

16.12

39.29

14.63

32.44

16.63

9.54

14.52

6.55

84.49

S3V2

32.64

52.54

6.48

19.67

-3.65

-3.80

4.87

-2.63

61.64

S1V3

19.11

20.36

17.53

24.08

3.38

5.58

27.06

0.77

78.63

S2V3

-13.65 1.22

0.49

9.42

2.76

6.67

-7.84

0.34

29.49

S3V3

-7.43

-5.17

-6.47

-1.15

-0.68

-3.47

-1.75

5.51

-4.69

Panel B Market adjusted returns MA

DI

MACD

STOCH

RSI

BB

CCI

NVI

SBH

S1V1

55.59

100.24

44.88

117.75

27.82

-7.59

22.38

40.34

157.43

S2V1

81.30

124.73

48.56

92.98

16.92

-18.71

36.40

10.84

117.81

S3V1

52.91

108.52

30.13

49.18

7.61

-9.05

2.26

9.19

107.37

S1V2

34.93

65.07

17.93

38.81

-6.26

-10.31

-5.10

-5.60

137.13

S2V2

-1.20

21.97

-2.69

15.12

-0.69

-7.78

-2.80

-10.77

67.17

S3V2

15.32

35.22

-10.84

2.35

-20.97

-21.12

-12.45

-19.95

44.32

S1V3

1.79

3.04

0.21

6.76

-13.94

-11.74

9.74

-16.55

61.31

S2V3

-30.97

-16.10

-16.83

-7.90

-14.56

-10.65

-25.16

-16.98

12.17

S3V3

-24.75

-22.01

-22.49

-23.79

-18.47

-18.00

-20.79

-19.07

-11.81

Decision, Vol. 32, No.1, January - June, 2005

112

Technical Analysis in the Indian Capital Market …A survey

Table 10 Profit (in %) for the year 2003 for different size and value portfolios Panel A Unadjusted Returns MA

DI

MACD

STOCH

RSI

BB

CCI

NVI

SBH

S1V1

222.22

249.54

203.39

278.96

31.35

16.63

56.90

47.74

302.70

S2V1

144.97

147.63

135.44

184.50

34.74

14.05

61.56

1.05

248.16

S3V1

148.55

127.85

131.72

213.01

69.13

19.79

70.87

2.85

296.48

S1V2

104.99

156.58

112.10

167.19

28.36

18.92

47.22

17.21

225.40

S2V2

146.11

136.74

103.87

190.91

34.67

15.48

52.16

17.09

215.35

S3V2

163.30

137.77

91.00

198.90

45.16

17.09

57.38

19.38

279.11

S1V3

91.41

95.00

67.63

139.37

66.25

26.98

94.90

12.91

182.52

S2V3

78.97

96.16

62.88

124.84

17.84

12.56

40.69

15.20

150.72

S3V3

18.88

39.74

26.85

53.24

13.51

5.23

8.93

5.23

52.42

SBH

Panel B Market adjusted returns MA

DI

MACD

STOCH

RSI

BB

CCI

NVI

S1V1

122.5

149.37

103.22

178.79

-68.82

-83.54

-43.27

-52.43

202.53

S2V1

44.80

47.46

35.27

84.33

-65.43

-86.12

-38.61

-99.12

147.99

S3V1

48.38

27.68

31.55

112.84

-31.04

-80.38

-29.30

-97.32

196.31

S1V2

4.82

56.41

11.93

67.02

-71.81

-81.25

-52.95

-82.96

125.23

S2V2

45.94

36.57

3.70

90.74

-65.50

-84.69

-48.01

-83.08

115.18

S3V2

63.13

37.60

-9.17

98.73

-55.01

-83.08

-42.79

-80.79

178.94

S1V3

-8.76

-5.17

-32.54

39.20

-33.92

-73.19

-5.27

-87.26

82.35

S2V3

-21.21

-4.01

-37.29

24.67

-82.33

-87.61

-59.48

-84.97

50.55

S3V3

-81.29

-60.43

-73.32

-46.93

-86.66

-94.94

-91.24

-94.94

-47.75

Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

113

If we check the status of recommended companies given in Table 3, the majority of the companies fall in S3V2 and S3V3. Even for the highly risk averse investor who wishes to concentrate on big and liquid stocks, there were better trading options available in S3V1 compared to the experts’ recommendations. For those investors who are willing to take some incremental risk owing to a shift in portfolio towards low and medium capitalisation stocks, the returns from value investing are even higher for most of the technical indicators. Hence, one can conclude from this that investors are ignoring strong return opportunities by not being cognizant of the impact of company characteristics on trading profits. However, as in Table 6, the empirical success of value based technical investing becomes questionable, once one compares the returns on technical analysis with SBH strategy. The superior performance of SBH strategy probably suggests that technical analysis is more of a myth than a reality. The empirical result suggests that TA being transaction extensive does not pay off after adjusting for transaction costs owing to frequent buying and selling of securities. Further, the period of evaluation witnessed a bull trend4 for the Indian market, making the sample share price series non-stationary over time. It is quite possible that technical indicators under evaluation are not appropriate for handling trending series. We recommend a more comprehensive evaluation for technical indicators that explicitly adjust for trending data vis-à-vis the SBH strategy. Further the empirical test should be conducted over a complete market cycle rather than for a given phase (up trend in our case). Tables 9 and 10(Panel B) are showing the results on double sorted portfolios net of market effects i.e. the return on a given portfolio minus the market return. The market return is defined as (Vt-Vt-1/Vt-1)*100 where Vt-1 and Vt are the values of BSE500 index at the beginning and end of calendar year. For estimating market adjusted %return, we should have preferably subtracted % return on the market factor from %return on each trade for a given sample stock before computing the annual stock returns and equally weighted portfolio returns. However, given the number of sample stock5, and the number of trades for each stock, and number of technical indicators, the computation would have been very heavy. Hence, adjusted returns have been approximated by subtracting the % annual return on market index from portfolio return. More reliable results can be obtained by empirically evaluating the technical trading system for different sets of stocks over a complete market cycle. Hence, our results can, at best, be described as suggestive. However, the survey findings throw light on the trading strategies of technical traders as well as the trading styles adopted 5.

Due to unavailability of data for some of the companies, a total of 373 and 422 companies out of 500 could be selected for finding out the return for the year 2002 and 2003 respectively.

Decision, Vol. 32, No.1, January - June, 2005

of a total

Technical Analysis in the Indian Capital Market …A survey

114

by them. Hence, they provide a platform for empirical researchers who wish to evaluate the economic feasibility of alternative technical indicators in the Indian environment using a representative study period. Summary and Concluding Remarks The survey covered institutional and individual respondents who have a long record of trading in the stock market and in the use of the technical trading system for securities’ analysis. The sample respondents believed that technical analysis could generate superior profits. They tend to apply these tools predominantly to the equity (and to some extent financial derivatives) segment of the market. A majority of them use TA along with the FA. Further, they tend to use technical analysis more frequently in the Bull Phase of the market. Surprisingly, the respondents revealed a greater preference for classical technical indicators, such as, trend analysis, gap analysis, candlestick charts, Fibonacci numbers and ratios, and simple moving average, while more sophisticated tools of TA seem to take a backseat. The respondents, further, felt that volume indicators provide independent information compared to price indicators and suggested that both of them played an important role in security selection. They were also not using time series econometric tools for TA. They employed, generally, the same tools for individual stocks as well as for market analysis. The stock recommendations provided by technical traders were related to big companies with high trading volume and medium P/B ratios. An empirical evaluation of size-value portfolios formed from the BSE500 index revealed that better returns could have been generated by investing in big and liquid companies with lower P/B ratios. For technical traders who will be willing to take extra risk, there are even stronger value bargains available in the small and mid capitalisation categories using selected technical indicators. However, the SBH strategy seemed to outperform all technical indicators for all categories of stocks, thus casting a shadow on the applicability of TA. However, more conclusive results can be obtained by performing such an evaluation over a complete market cycle. The survey findings shall be extremely relevant for technical traders who are continuously on the look out for investment strategies that beat the market. They shall also be useful for financial software developers who wish to know what they should emphasise in the technical toolbox. They are also relevant for empirical researchers who intend to test the applicability of technical tools. Finally, it has implications for market regulators and the investing community in general, as it shall help, in a better way, the understanding of investor behaviour, in the short run, for the Indian market.

Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

115

References Alexander, E. 1993. Financial trading. London: Kogan Page Limited. Alexander, S. S., 1961 .Price movements in speculative markets: Trends on random walks. Industrial Management Review, 2 (May): 7-26. Alexander S. S.,1964. Price movements in speculative markets: Trends on random walks, No.2. Industrial Management Review, 5, Spring: 25-46. Andrew, W., Lo, Mamaysky, H and Wang, J 2000. Foundations of technical analysis: Computational algorithms, statistical inference, and empirical implementation, The Journal of Finance, 55 (4) :1705-1765. Batten, and Ellis C., 1996. Technical trading system performance in the Australian share market: Some empirical evidence, Asia Pacific Journal of Management, 13 (1) : 87-94. Bhole L.M. 2002. Financial institutions and markets. 3rd Edition. New Delhi: Tata McGraw Hill Publishing Company Ltd. Blume, L, Easley, D and O’Hara, M. 1994. Market statistics and technical analysis: The role of volume. The Journal of Finance, 49 (March): 153-181. Brock, Lakonishok J and Le Baron, B. 1992. Simple technical trading rules and the stochastic properties of stock returns. The Journal of Finance, 47(5) : 17311764. Cootner, P. H.1962. Stock prices: Random vs. systematic changes. Industrial Management Review, 3, (Spring): 24-25. Dawson, S. M., 1985 Singapore share recommendations using technical analysis. Asia Pacific Journal of Management, 2(3): 180-188. Edward, R.D. and Magee, J. 1996. Technical analysis of trends, 4th Edition, Springfield, Massachusetts: John Magee. Fama, E. F., and Blume, M. E. 1966. Filter rules and stock market trading. Journal of Business, 39 (Special supplement – January): 226-241. Gujarati, D. N. 1995. Basic Econometrics. 3rd Edition. Singapore: McGraw Hill, International Editions. Fogler,H. R. 1978. Analysing the stock market: Statistical evidence and methodology, 2nd Edition. Grid Inc.: Columbus, Ohio.

Decision, Vol. 32, No.1, January - June, 2005

Technical Analysis in the Indian Capital Market …A survey

116

James, F.E. Jr. 1968. Monthly moving averages – An effective investment tool? Journal of Financial and Quantitative Analysis, (Sept.): 315-326. Jobman, D.R. 1995. The handbook of technical analysis. Burr Ridge, Illinois: Irwin Professional Publishing. Lakonishok, J., Shleifer, A. and Vishny, R. 1994. Contrarion investment, extrapolation and risk. The Journal of Finance, 49: 1541-1578. Levy, R. A., 1967. Random walks: Reality or myth. Financial Analysts Journal, (November-December): 69-77. Mitra S. K. Profiting from technical analysis in Indian stock market. Finance India, 16 (1) : 109-120. Neftei, S.N.and, Policano. A. J.1984. Can chartists outperform the market? Market efficiency tests for technical analysis. The Journal of Futures Market, (4) : 465-478. Plummer, T. 1989. The psychology of technical analysis. Chicago, Illinois: Probes Publishing Company. Pring, M.J. 1991. Technical analysis explained. 3rd Edition. New York: McGraw Hill. Sehgal, S. and Garhyan, A. 2002. Abnormal returns using technical analysis: The Indian experience. Finance India, 16 (1) : 181-203. Sullivan, R. Timmermann, A. and White, H. 1999. Data snooping, technical trading rule performance and the bootstrap. The Journal of Finance, 54 (5) : 16471691.

Decision, Vol. 32, No.1, January - June, 2005

117

Technical Analysis in the Indian Capital Market …A survey

Appendix TECHNICAL ANALYSIS IN INDIAN CAPITAL MARKET

Please mark an asterisk (*) against the option you wish to select. For example, to select Yes, mark as follows:

Yes: *

No:

I.

PERSONAL PROFILE

1)

Name:

2)

Investor Category

3)

Your experience in the stock market (in years):

4)

Your experience in Technical Analysis (in years):

II.

TECHNICAL ANALYSIS - INDIVIDUAL SECURITIES

1)

Institutional:

Do you believe Technical Analysis helps in generating better returns? Yes:

2)

Individual:

No:

Do you use Fundamental Analysis along with Technical Analysis? Yes:

No:

3)

Name the technical Software(s) used (if any):

4)

How extensively you are using Technical Analysis in different stock market phases? (Please rank on a 5 point scale. 1 for Not using and 5 for Extensive use.)

Decision, Vol. 32, No.1, January - June, 2005

118

Technical Analysis in the Indian Capital Market …A survey

Bull Phase: Bear Phase: Normal Phase: 5) Please select the relevant category/categories in which you mostly use technical indicators. (a)Shares: (b) Bonds: (c)Financial Derivatives: (d) Forex: 6) Do you think that the holding period of technical recommendations is influenced by the nature of settlement systems (such as Rolling Settlement, Weekly Settlement, etc.) on the stock exchanges? Yes: No: If yes, please specify, the reasons: 7)Do you believe that returns on the shares using technical tools get affected by circuit breakers and circuit filters? Yes: No: 8)What scale do you use in your charts? Arithmetic scale: 9)

Logarithmic Scale:

Both:

Please mark asterisk (*) against the technical tool/indicator you normally use during different market phases. Also indicate the extent of your preference for these indicators on a 5-point scale. (1 for no preference, 5 for the highest preference). Also specify the windows (number of days) used for different technical indicators e.g. 12-day moving average, a 7- day momentum etc. Bull Period Mark* Prefer((If ence used) (1-5)

Bear Period Mark* ((If used)

1.Trend lines 2.Gap Analysis 3.Japanese candlesticks 4.Elliot Wave 5.Fibonacci numbers 6.Moving Average 7. Moving Average convergence divergence 8.MACD-Histogram 9.Directional System (ADX)

Decision, Vol. 32, No.1, January - June, 2005

Preference (1-5)

Trading Range Mark* ((If used)

Preference (1-5)

Windows Used (no. of days)

119

Technical Analysis in the Indian Capital Market …A survey 10. Momentum 11. Rate of Change (ROC) 12. William %R 13. Stochasic 14.Relative Strength Index (RSI) 15.On Balance Volum (OBV) 16. AccumulationDistribution 17. Open Interest 18.Commodity Channel Index (CCI) 19. Force Index 20. Elder Ray 21. Signals from the Press 22. Insider Trading 23. Odd-Lot Activity Any other (Please specify) 24. 25.

10)

Select the trading systems, if any, used by you. (a) (b) (c) (d)

Triple Screen Trading System: Parabolic Trading System: Channel Trading Systems: Any Other:

If Triple Screen Trading System is used, please specify the technical indicators preferred in the-

11)

First Screen: Second Screen: Are you using price indicators, volume indicators or both? Price Indicators: Volume Indicators:

Decision, Vol. 32, No.1, January - June, 2005

Both:

120

Technical Analysis in the Indian Capital Market …A survey 12)

Indicate your opinion for the following statements relating to the use of volume and price indicators: (i) Volume indicators are preferred over price indicators. Yes: No: (ii) Volume indicators lead price indicators. Yes: No: (iii) Volume indicators and price indicators are related (consistently give similar signals). Yes: No: (iv) Price indicators are confirmed by volume indicators. Yes: No:

13)

Do you also use time series models with technical analysis (such as ARIMA, ARCH, GARCH processes etc.)? Yes: No:

14)

Name the stocks on which you regularly fit the technical tools with high success rate in the recent past and also provide their average holding period in days. Stock

Holding period

Stock

i.

vi.

ii.

vii.

iii.

viii.

iv.

ix.

v.

x.

Holding period

III.

TECHNICAL ANALYSIS - MARKET, SECTORAL AND COMPANY CHARACTERIS TICS RELATED PATTERNS

1)

Put an asterisk (*) against the index that you use as a barometer of stock market? SENSEX BSE100 BSE200 BSE500

S&P CNX Nifty

NSE

Any other (Specify)

2)

Do you use the same set of technical tools for analysis of individual stock as well as market index?

3)

Do you use the following tools/indicators for market index?

(a)

New High New Low (NH-NL):

Yes:

(b)

Advance Decline Ratio:

(c)

Trader’s Index (TRIN):

(d)

Any Other:

No:

Decision, Vol. 32, No.1, January - June, 2005

121

Technical Analysis in the Indian Capital Market …A survey Technical Analysis in Indian Capital Market …A survey 4)

Mark an asterisk (*) against the industrial sectors in which you are actively using technical tools. IT

Telecom

Entertainment

Pharmaceutical

Banking

FMCG

Cement

Steel

Refineries

Fertilizers

Power

Food Products

Capital Goods

Other industries

5)

Do you use a different set of technical indicators for the analysis of the following categories of stocks? Indicate by marking an asterisk in the respective column. Ye s A.

Small stocks and big stocks

B.

High P/E ratio and Low P/E ratio

C.

High BE/ME and Low BE/ME

D.

High Dividend yield and Low Dividend yield

E.

High Leverage and Low Leverage

F.

Old economy and New economy shares

G.

Index and Non Index stocks

H.

High turnover and Low turnover stock

Decision, Vol. 32, No.1, January - June, 2005

No

Technical Analysis in the Indian Capital Market …A survey

6)Please specify the preferred tools/ indicators for the stock categories for which the answer in the previous question is in the affirmative. Company Characteristics A.

Name of the tool/indicator used

Big stocks Small stocks

B.

High P/E stocks Low P/E stocks

C.

High BE/ME stocks Low BE/ME stocks

D.

High Dividend stocks Low Dividend stocks

E.

High Leverage stocks Low Leverage stocks

F.

Old economy shares New Economy shares

G.

Index stocks Non Index stocks

H.

High turnover stocks Low turnover stocks

Thank you very much for your precious time.

Decision, Vol. 32, No.1, January - June, 2005

122