Comparison of technical indicators

ABSTRACT

Trading in today's markets requires a fair amount of skill, fundamental knowledge of finance, and good technical expertise in judging and predicting the latest market moves. Technical analysis of financial instruments is an integral part of successful trading. With numerous technical indicators available to the modern trader, it can sometimes get quite tricky figuring out which indicator best suites one's style and produces the best results for specific financial instruments. Even though many indicators have been extensively researched and agreed on, full reliance on their indications is still a matter of concern.

This report proposes a systematic and automatic approach to technical analysis of the SGX stock market through a comprehensive study and comparison of the different technical indicators. Through numerous analyses on various stocks and commodities, this report presents a clear perspective on how accurately these technical indicators work and how much profit/loss they generate. The best features of these indicators are incorporated as rules in a trading system, and then implemented in Metastock as an Expert Advisor. This system is tested on historical financial data to check its feasibility in the SGX stock market and fine-tuned to integrate the different market parameters and trading formulae. Eventually, this Expert Advisor will provide any Metastock user with a smart strategic approach towards trading in today's financial markets.

ACKNOWLEDGEMENT

This project and report would not have been possible if not for the help and support of many people. I would like to express my sincerest gratitude to all that have helped me achieve the goals of my Final Year Project.

First and foremost, my sincerest to Dr. Patricia Wong Jia Yiing, Associate Professor, School of EEE, for giving me the opportunity to work on a project which I was very interested in taking up. Her guidance in defining the scope of my project was of immense help and it set me on track for the rest of the duration of the project. Her judicious thinking and attention to detail inspired me. I have learnt a lot from her, and am really grateful for her time and effort, and also her invaluable career advice.

I would also like to take this opportunity to thank some of my colleagues at Credit Suisse, Singapore (where I have interned), whose advice and knowledge helped me understand the financial world and trading in the stock markets.

INTRODUCTION

BACKGROUND

"The statement "market action discounts everything" forms what is probably the cornerstone of technical analysis. The technician believes that anything that can possibly affect the price--fundamentally, politically, psychologically, or otherwise--is actually reflected in the price of that market [1]." - John J. Murphy (1999)

Technical analysis is the study of past security prices and volumes, while attempting to predict the future. This study only deals with the price elements (high, low, open and close) and the volume associated with a security. Other factors such as earnings, revenue growth, etc. are ignored. Technical analysis can be applied to just about any investment vehicle, including stocks, futures, options, or mutual funds. Traders and technical analysts employ many rules and financial models based on price and volume transformations, such as moving averages, regressions, indexes, inter-market and intra-market price correlations, cycles or, classically, through recognition of chart patterns. For this reason, this approach is also known as 'charting'.

Technical analysis is a very successful method of trading, mainly because it removes the 'emotional' outlook of the trader. Decisions are taken based solely on specific trading rules and indications. For the implementation of these indicators as technical systems, many financial models and software are available, including Metastock by Equis International. This software has been used for the purpose of this report.

Successful trading in today's markets requires also a fundamental analysis of the financial instruments; an in-depth study of the different market moves and the financial and economic situations. Equally important is the technical analysis. There are over 50 common and successful trading indicators available for the common man today. Some of these grow outdated with time, while some adapt to provide better results. In the following sections, a detailed analysis of some of the most famous indicators, along with their application and results on the SGX stock market, is provided. While different indicators give buy/sell signals in different ways, their most successful components are integrated into a trading system implemented as an Expert Advisor in the Metastock software.

It is not possible for any trading system to forever come up with successful results. There are bound to be a few loose trades and losses. However, a good trading system is one which can quickly identify downward movements and provide good alerts and signals to recover one's positions. Identifying the trend is very essential, as is knowing the right time to exit a position. All in all, a good trading system is one which best limits the losses while providing good substantial profits in the long run.

PROJECT SCOPE AND OBJECTIVES

SCOPE

The scope of this project is to study and compare the different technical indicators, understand their working principles and examine their validity in today's financial markets. These indicators shall be used solely or combined together to form new trading systems in Metastock. Through extensive testing of these systems on historical financial data, this report provides a precise idea about these indicators and their biggest advantages and disadvantages. The best features of these indicators are then implemented in an elaborate trading system that integrates all market features - trend, momentum and volume. This system shall work as an Expert Advisor and provide the user with all the necessary tools and information required for successful trading in the SGX stock market.

OBJECTIVES

There are three major objectives of this project:

  1. To learn about technical analysis of stocks and other financial instruments.
  2. To study and compare the various technical indicators by testing them on SGX stock data.
  3. To develop a trading system in Metastock using Metastock programming and testing its performance on historical data.

The system thus developed shall then be implemented as an Expert Advisor, providing the user with a set of trading rules to trade objectively on the stock market.

PROJECT APPROACH AND TIMELINES

PROJECT APPROACH AND BIBLIOGRAPHY

The process through which the objectives of this project were completed is as explained below:

The theoretical knowledge of technical analysis and indicators was derived from numerous sources, mainly books and internet. Some books which proved especially useful were:

  • Technical Analysis of Financial Markets by John Murphy
  • Technical Analysis of Stock Trends by Robert.D.Edwards & John Magee
  • Getting Started in Technical Analysis by Jack Schwager
  • Technical Analysis Explained by Martin.J.Pring
  • Technical Analysis from A to Z by Steve Achelis.

PROJECT TIMELINES

Thus, it took roughly 35 weeks to complete the whole project. This report shall describe the different stages, the studies and research, and the approach taken for developing the Expert Advisor.

LITERATURE REVIEW

TECHNICAL ANALYSIS

Technical analysis is the process of identifying price patterns andtrendsin financial markets in an attempt to exploit those patterns. Various methods and tools are used for this purpose but the most essential is the study of charts. Technical analysts then employ indicators on the chart data to better interpret and analyze the market movements. These indicators are inherently mathematical transformations of price or volume and give a general indication of the security based on its trend, volume or momentum. These indicators also attempt to determine the price direction of the security. Some commonly famous technical indicators include:

  • Moving Average Convergence Divergence
  • Stochastic Oscillator
  • Money Flow Index
  • Relative Strength Index
  • Accumulation/Distribution

An in-depth analysis of these indicators is provided in the following section. Systems were developed for these indicators, tested on SGX stocks and their results compared and evaluated.

Before applying these technical indicators to the stock data, it is essential to understand the fundamentals of technical analysis.

CHARTS

Technical analysis, also known as 'charting' is essentially the study of charts. Charts display the price history of a security, the most important information for a trader. From this the trader can assess the market volatility and the risk involved in taking up any positions. Even in fundamental analysis, charts are very significant as they enable the analyst to determine the periods of major price moves. By identifying the market changing events in those periods, one can get a good idea of the price influencing factors. This information can then be used to construct a price behavior model. Charts are also frequently used as timing tools, even by analysts who depend on other information for their trading decisions.

The ability to study charts is very vital for technical analysis as they are the most important money management tool. They help to define meaningful and realistic stop points. Charts also reflect certain repetitive patterns in the market behavior of a security. A technical analyst can easily anticipate market and price moves through a quick study of the chart. An understanding of charts is hence a prerequisite for developing profitable technical stock trading systems.

TRENDS

The most common thing a technical analyst looks for in a chart is trends, i.e. a succession of higher highs or higher lows. An uptrend is defined is a period during which each successive high is higher than the preceding highs. Similarly, a downtrend is when each low is lower than its preceding lows. A trend is considered intact until a previous reaction point is broken.

This condition implies that the trend is broken and serves a warning to the trader to reconsider his positions. However, it should only be considered to be an indication and not a sure-shot sign of a long term trend reversal. There are many other ways of confirming the origination and end of a trend, mainly by simultaneously checking the volume and momentum of the market. This is usually achieved by technical analysis using different indicators and avoiding multi-co linearity. It should be noted that trends also exist in the horizontal direction, i.e. a period of no higher highs or lower lows but only normal trading with no dramatic price moves.

Up-trends and down-trends are often highlighted with trend lines, which make it easier to see how long the trend exists.

Trends can be most successfully analyzed in the following ways:

  • Positions may be initiated in the direction of the major trend when an uptrend is approached by a decline or a downtrend with a price rally.
  • The break of a uptrend is a symbol for selling and that oof a downtrend is a signal for buying. A break should be confirmed by a minimum price change percentage or a minimum number of price moves beyond the trend line.
  • The lower end of a downtrend channel and the upper end of a uptrend channel represent potential profit points for short term traders.

Many traders over-estimate the reliability of trend lines by establishing them in hindsight. It is essential that a trend line is re-drawn as a bull or bear market is extended. This ensures that although the penetration of the trend line offers a warning signal of a trend reversal, this development can be easily accounted for by a mere re-drawing of the trend line.

SUPPORT AND RESISTANCE

Support and resistanceis a concept intechnical analysisthat says that the movements in the price of asecuritystop and reverse at certain predetermined price levels.

Support: A support level is the price at which buyers are expected to enter the market in sufficient numbers to takecontrolfrom sellers [2].

The market has a memory. When price falls to a new low and then rallies, buyers who missed out on the first trough will be inclined to buy if price returns to that level. Afraid of missing out for a second time, they may enter the market in sufficient numbers to takecontrolfrom sellers. The result is a rally, reinforcing perceptions that price is unlikely to fall further and creating a support level [2].

Resistance: A resistance level is the price level at which sellers are expected to enter the market in sufficient numbers to takecontrolfrom buyers [2].

When price makes a new high and retreats, sellers who missed the previous peak will be inclined to sell when price returns to that level. Afraid of missing out a second time, they may enter the market in numbers sufficient to overwhelm buyers. The resulting correction will reinforce market perceptions that price is unlikely to move higher and establish a resistance level [2].

Once the support levels are penetrated, they often become resistance levels andvice versa. This can be explained by simple market logic: buyers, who purchase near a support level, and then realize that the price has fallen, will now want to sell in order to recover their losses, when price rallies to reach their break-even point. The support level thus becomes a resistance level. Likewise, traders who sell when price approaches a resistance level will be let down by a market move beyond the trend line and a continuous rise of price. They will now want to buy if price returns nearby the support level, so that they do not miss out a second time. The resistance level is now redefined as a support level.

MOVING AVERAGES

Moving Averages are a very common and fruitful method of technical analysis. Many technical indicators are derived from the principles of moving averages since they provide an objective measure of trend direction by smoothing price data. Normally calculated using closing prices, the moving average can also be used withmedian,typical,weighted closing, and high, low or openprices as well asother indicators.

Every type of moving average is a mathematical outcome of an average of a number of past data points. Once calculated, the resulting average is then plotted onto a chart in order to allow traders to look at the smoothed data rather than focusing on the day-to-day price fluctuations that are inherent in all financial markets.

The number of past data points selected for calculation depends on the time frame. Shorter length moving averages would be more sensitive and are suitable for identifying new trends quickly, but they also end up giving more false alarms. Longer moving averages are more reliable but less responsive since they only pick the big trends.

Moving Averages may be taken in different ways. Some of these are explained below:

Simple Moving Average: It takes the un-weightedmeanof the previousndata points.

Weighted Moving Average: It is any average that has multiplying factors to give different weights to different data points. These weights decease arithmetically.

Exponential Moving Average: It applies weighting factors which decrease exponentially. By definition, the weighting for each older data point decreases exponentially, giving much more importance to recent observations while still not discarding older observations entirely.

The simplest moving average system generates signals through crossovers, i.e. when price crosses the moving average or the lower moving average crosses over/below the higher moving average. In general the rules for moving average crossovers are:

  • Go long when price crosses to above the moving average from below or a lower moving average crosses above a higher moving average.
  • Go short when price crosses to below the moving average from above or a higher moving average crosses below a lower moving average [3].

The system based on moving averages would be prone to a lot of movement and penetrations of the trend line in ranging markets, with price crossing back and forth across the moving average. This would generate a large number of false signals, commonly known as whipsaws. For this reason, moving average systems normally employfiltersto reduce whipsaws.

The popularMACDindicator is a variation of the two moving average system, plotted as an oscillator which subtracts the slow moving average from the fast moving average.

COMMON CHART PATTERNS

Achart patternis apattern formed within achartwhen prices are arranged as graphs. Instock market trading, chart pattern studies play a large role duringtechnical analysis. Whendatais plotted technical analysts look for patterns which occur and repeat themselves over a period of time. Chart patterns are used as either reversal or continuation signals.

Some commonly occurring chart patterns are explained below. These were made common by Edwards and Magee and are widely used by traders and investors.

Head and shoulders: Thehead and shoulders patternis a frequently foundpatternin the price charts of financial instruments (stocks, bonds, futures, etc). The pattern derives its name from the fact that its shape is visually similar to that of a head with two shoulders. It is one of the most reliable chart patterns, and it reaches its projected target in 95% probability [4].

As illustrated in the figure above, a head and shoulders pattern has four basic elements: a left shoulder, which is essentially a relative peak; a head, which is a peak above both shoulders; a right shoulder, which is a peak that is approximately parallel to the left shoulder; and a neckline, which forms a support, or bottom, of price activity within the range of the head and shoulders pattern [4].

Since the necklines serves as a support, traders often look to enter a position on the break of this trend line. Upon this break, traders often look to set a target profit equal to the distance between the neckline and the head of the chart. Astop loss orderis frequently set at where the right shoulder is. As a result, the head and shoulders pattern can be used by traders not only to identify entry points, but also to manage the risk of the trade [4].

Cup and Handle: Thecup and handleformation is another common pattern found in stock and commodities trading. It is a bullishchart pattern defined by a chart where a stock dips in value, rises back up to the original value, then dips a small amount in value, and then rise again by a small amount in value.

Ideally, a cup and handle formation should follow an increase trend, generally about a few months old. Also, the cup must always precede the handle. The cup should form a rounded bowl shape, with an obvious bottom. Most analysts believe a V shaped bottom should be avoided.The cup should be fairly shallow. Most cup patterns retrace about 30% to 50% of the previous increase. The perfect pattern would have equal highs on either side of the cup, but that is very hard to find and is a rare occurrence.

The cup should last for a few months, while the handle should only last for 1 to 4 weeks.In most cases, it is observed that the cup pattern continues over years followed eventually by a short handle. Another confirmation of this pattern is that the volume of the stock should decrease along with the price during the cup and increase rapidly near the end of the handle when the price begins to rise. Most traders consider a cup and handle formation to be abullishsignal since it is usually followed by a sharp rise in value.

Triangles: Trianglesare another commonly found pattern shapes in the price charts of financial instruments. It is characterized by a narrowing in price range and converging trend lines, thus giving it a triangular shape and getting its name.

Triangle Patterns are categorized into three categories: the ascending triangle, the descending triangle, and the symmetrical triangle. The shape of the triangle is significant, but of more importance is the direction that the market moves when it breaks out of the triangle [5]. Triangles are normally seen as continuation patterns but they can also signify trend reversals in a few cases.

The ascending triangle patterns are normally seen in an uptrend and treated as a continuation pattern as buying increases, continuing up to the top resistance line of the pattern. If this pattern is observed in a downtrend, it should be considered as a trend reversal. Similarly, descending triangle patterns are normally seen in a downtrend and treated as a continuation pattern as selling increases, continuing down to the bottom support line of the pattern. And again, this pattern signifies trend reversal if observed in an uptrend.

Double top and double bottom:Adouble topis a reversalchart pattern defined by a chart where the price of afinancial instrument increases up to a particular level, then drops back from that level, increases again to that level, and then finally drops back off again. A double bottom is the exact opposite.

As shown above, in the Double Top formation, two peaks are formed and in the double bottom, two v shaped valleys. These patterns form easily in rapidly increasing or decreasing markets. Manyanalystsoften misread and misinterpret these signals. At the second shape, there is no confirmation pointing to Double Top or Bottom and mostly thetrendremains in force and prices simply move up to make new highs or lows. An essential condition here is that the Double Top or Bottom should be shaped over a few months and not very close together. The formation should also be backed up by relevant volume movements. For instance, in the formation of a double top, the volume should gradually decrease with the fall of the first peak and increase to the second peak with reduced volume. This is an important confirmation for the formation of these chart patterns.

Flags and Pennants: Theflag and pennant patternsare also commonly foundpatternsin price charts. These patterns are characterized by an apparent direction of the price trend, followed by a consolidation and range bound movement, which is then followed by a resumption of the trend.

As illustrated above, the flag pattern is encompassed by two parallel lines. These lines may be flat or pointed in the opposite direction of the primary trend in the market. The pole line then represents the primary trend of the market. The pattern basically signifies that the market might be relaxing after a big move and shall continue its primary trend.

The pennant pattern is identical to the flag pattern in its shape and significations; the only difference being that the consolidation phase of a pennant pattern is highlighted by converging trend lines rather than parallel trend lines.

Wedges: This pattern is characterized by a contracting range in prices together with a rising trend in prices (called a rising wedge) or a downward trend in prices (called a falling wedge).

A wedge pattern is considered to be a momentary arrest in the movement of a primary trend. In this type of formation, trading activities are confined within converging straight lines, hence forming a pattern. It generally takes about 3 to 4 weeks to complete the wedge. This pattern has a rising or falling slant pointing in the same direction [6].

Wedges differ from the triangle regarding the boundary lines, which either slope up or down in the formation of a wedge.The pricebreaking out point forms another difference from the triangle. Falling and rising wedges are a small part of an intermediate or major trend. As they are reserved for minor trends, they are not considered to be major patterns. Once that basic or primarytrendresumes itself, the wedges lose their effectiveness as technical indications [6].

TECHNICAL INDICATORS

Various technical indicators were researched for the development of this project. Systems were developed and tested for these indicators on the SGX stocks. The results were compared and evaluated to come up with an eventual system which gives the most profitable results.

MOVING AVERAGE CONVERGENCE DIVERGENCE (MACD)

The Moving Average Convergence Divergence (MACD) is a technical analysis indicator created by Gerald Appel in the 1960s. It is based on the difference between a fast and slow exponential moving average (EMA) of closing prices. As explained earlier, moving average (MA) is a mathematical procedure that records the average value of prices or other data over a specific period of time. Because they integrate a stream of these average values, MAs smooth out a data series and make it easier for the analyst to spot trends. This makes MACD a trend following indicator. A major disadvantage of using moving averages is that they are lagging indicators.

The standard MACD takes the difference between a security's closing price's 26-day and 12-day Exponential Moving Averages (EMAs). Another 9-day EMA is plotted along with these to provide a trigger line. This indicator is calculated as:

MACD = EMA [12] - EMA [26]

Signal = EMA [9] - MACD

The difference between the MACD and the signal is usually shown in the form of a histogram, which enables easy indication of buy/sell signals. A bullish crossover is said to occur when the MACD moves above its 9-day EMA and a bearish crossover when the MACD moves below the 9-day EMA. As the figure below shows, the histogram is positive when MACD is above its 9-day EMA and negative when MACD is below its 9-day EMA.

Here, the 12 day EMA moves faster than the 26 day EMA giving a MACD trend line which is compared with a 9 day EMA.

There are 3 common methods of using MACD to generate buy/sell indicators:

Moving Average Crossover: It indicates 'buy' when the MACD crosses up through the signal line and 'sell' when the MACD goes below the signal line.

Center-line crossover: A MACD crossover above the zero line is interpreted as bullish and a crossover below the zero line as bearish.

Divergence: A trend may be indentified to be beginning or ending through a divergence in the MACD and the security price.

The MACD produces the most successful results in strongly trending markets. This indicator also highlights overbought and oversold conditions. An overbought situation occurs when prices rise too far too fast and are ready for a downward correction. An oversold situation occurs when prices fall too far too fast and are ready for an upward correction. When the shorter moving average pulls away from the longer moving average (i.e., the MACD rises), it is likely that the financial instrument's price is too high and will soon return to more realistic levels [7].

An indication to the end of the current trend occurs when the MACD diverges from the financial instrument's price. A negative/bearish divergence is said to occur when the MACD is making new lows while prices fail to reach new lows. Conversely, a positive/bullish divergence is said to occur when the MACD is making new highs while prices fail to reach new highs. Both of these divergences become even more significant when they occur at relatively overbought/oversold levels [7] in the stock market.

STOCHASTIC OSCILLATOR (SO)

This indicator was developed by George Lane in the 1950s and is used to indicate momentum by measuring the currentclosingpricerelative to the high/low range of a defined period, the most common of which is 14 days. It best compares the current closing price relative to the high/low range over a set number of periods [8]. Buying pressure is indicated by closing levels near the top of the range and selling pressure by those near the bottom.

This indicator is calculated as:

STS = 100 x (closing price-price low)(price high-price low)

A 14 day %K (14 day stochastic oscillator) uses the most recent close, the highest high and the lowest close of the previous 14 days. The defining theory of the SO indicator is that prices tend to close near their high for upward trending markets, and near their low for downward trending markets. Transaction signals are expected when the %K crosses through a three-period moving average called the "%D".

There are three well established methods for using the %K and %D indicators to decide when to buy/sell stocks.

  • In the first case, %D acts as a trigger or signal line for %K. A buy signal occurs if %K crosses up through % D and a sell signal if it crosses down through % D. Such crossovers usually occur too often. To avoid whipsaws and other drawbacks, one should look for other indicators as well such as waiting for crossovers to occur together with an overbought/oversold pullback, or after a peak or trough has formed in the %D line. A SMA of the %D indicator may also be taken if price volatility is too high. This efficiently smoothes out rapid fluctuations in price.
  • In the second case, %K or %D levels above 80 and below 20 may be implemented as overbought or oversold conditions. This basically implies that one should buy or sell after a bit of a reversal and once the price crosses one of these threshold limits, the investor should wait for prices to return through those thresholds. This is the best way of implementing this indicator, along with confirmations from other indicators.
  • The third way that this indicator might be used is to watch for divergences where the Stochastic trends in the opposite direction of price. This is a strong indication that the momentum in the market is declining and there might be a reversal in the near future.

MONEY FLOW INDEX (MFI)

The Money FlowIndex (MFI) is a volume-weighted indicator, and hence enables a quick measure of the amount of money flowingin and out of a security. This indicator uses money flows to compare prices and identify the strength or weakness of a trend. It is measured on a 0 - 100 scale and is often calculated using a 14 day period [9]. In many aspects, it is similar to the momentum measuring indicator Relative Strength Index (RSI) and hence incorporates momentum as value as volume in providing signals.

The terminologies and formulae of this indicator are as follows:

Typical Price = High+Low+Close3 ; Money Flow = Typical Price x Volume

Money flows are defined as positive and negative. Positive money flowis the total for those days when the typical price is higher than the previous day's typical price, andnegative money flowwhen below.

Money Ratio = Positive Money FlowNegative Money Flow

Money Flow Index = 100 - 100(I+money ratio)

The MFI generates buy/sell signals through:

Divergence: If the stock price is falling but the positive money flow is still greater than the negative money flow, this implies that there is more volume associated with the daily price rises than the price drops [9]. This indicates that the trend is weak and will reverse if the money flowing into the security is more or stronger than the money flowing out of it.

Overbought/Oversold conditions: The MFI is used to indicate the quantity of volume associated with a security. As shown in the figure above, stock would be considered to be overbought (bearish) if the MFI indicator reaches 80, and oversold (bullish) if it falls below 20. This is the most common and successful method of implementing this indicator, along with confirmations from other indicators.

RELATIVE STRENGTH INDEX (RSI)

The Relative Strength Index (RSI) is a technical analysis oscillator which shows price strength by comparing upwards and downward close-to-close movements. This indicator was developed by J. Welles Wilder and introduced in his 1978 book, New Concepts in Technical Trading Systems. Since its introduction, the RSI has been an extremely useful and popular momentum oscillator.

This indicator compares the magnitude of a stock's recent gains to the magnitude of its recent losses and converts that information into a value ranging from 0 to 100. This calculation helps to determine overbought and oversold conditions of an asset. Here only one parameter is required, the number of time periods to use in the calculation.

For each day an upward change (U) or downward change (D) is calculated. 'Up' days are those for which the daily close is higher than the day before's daily close, i.e.:

U = closetoday - closeyesterday , and

D = 0

Conversely, a 'down' day is when the close is lower than the previous day's. It should be noted that D would still be a positive number.

U = 0, and

D = closeyesterday - closetoday

In the case of today's close being the same as yesterdays, both U and D values would be zero. An average of U values is calculated with an exponential moving average using a given N-days smoothing factor, and the same for D [10]. The ratio of those averages is defined as the Relative Strength.

Relative Strength (RS) = EMANof UEMANof D

This is then converted to a value between 0 and 100 as:

Relative Strength Index (RSI) = 100 - 100(1+RS)

This RSI formula may also be written as follows:

RSI = 100 * EMANof UEMANof U+ (EMANof D)

This formula better emphasizes the way RSI expresses the up as a proportion of the total up and down. Generally a smoothing time period of 14 is used (as initially proposed by Wilder).

A security is considered overbought if the RSI reaches the 70 level. This implies that the trader should consider selling. Conversely, the security is considered oversold at the 30 level. This is based on the principle that when there's a high proportion of daily movement in one direction it suggests an extreme, and prices are likely to reverse. The centerline for RSI is 50. Readings above and below can give the indicator a bullish or bearish tilt [10].

Buy and sell signals are also generated by looking for positive and negative divergences between the RSI and the underlying security. For example, consider a falling stock whose RSI rises from a low point of (say) 15 back up to about 55. Because of how the RSI is constructed, the underlying security will more often than not reverse its direction soon after such a divergence. As in this example, divergences that occur after an overbought or oversold reading usually provide more reliable signals [11].

It is observed that most traders who use the RSI focus their attention on trying to identify bullish and bearish divergences [11]. Basic price and momentum divergences play an important role in identifying extreme overbought or oversold conditions in the market.

Most traders fall prey to the concept of divergence and see it as the end or reversal of the existing trend of the market. However it often occurs that sentiment and momentum keep the market continuing to make new highs (or lows). This keeps the RSI at overbought/oversold levels for extended periods of time [11].

Momentum and price corrections, when they do appear, are generally sharp and quick. After these brief respites the market then resumes its normal upward/downward trend. With each successive new high/low and divergence formed, concerned traders attempt to call for a top/bottom and reversal of trend. However, multiple divergences frequently develop in strongly trending markets and this only leads to corrections of the overbought (oversold) condition of the market. This is a major reason why divergences are not mostly used as buy/sell signals for this indicator. The RSI is best used in conjunction with other indicators.

ACCUMULATION / DISTRIBUTION INDEX

The Accumulation / Distribution indicator helps to track the relationship between price andvolume. It is a leading indicator of price movements and provides a measure of thededicationof the bulls and bears to the market. This indicator can also be used to detectdivergencesbetween volume and price action - a major sign that a trend is weakening.

The Accumulation / Distribution works according to the following formulae. First, a Close Location Value (CLV) is calculated as:

CLV = close -low- (high -close)(high -low)

This value of CLV ranges from -1 (when the close is the low of the day), to +1 (when the market ends at the high of the day). The accumulation/distribution index then adds up the volume multiplied by the CLV factor:

A / D = A / D previous + volume x CLV

In general, the shape of the Accumulation/Distribution index is more important than the actual value. The name Accumulation/Distribution comes from the fact that during accumulation buyers acquire control over the market and the price bids up through the day, or makes a recovery if sold down. It is hence seen that the price and the A/D value more often finish near the day's high than the low. The opposite occurs for distribution.

The strongest signals for the Accumulation / Distribution indicator are obtained through divergences. A positive/bullish divergence occurs when the A/D indicator rises in value even when the price is falling. This signifies that the trend is not very healthy and will most likely die out soon. If the stock price is falling rapidly but theAccumulation/Distribution index Line is not making lower lows or is trading sideways, it serves as an indication that buying pressure is reasonably weak. Similarly, a negative/bearish divergence occurs if the stock price is increasing but this movement is not substantiated by an upper movement in the Accumulation/Distribution index. If the Accumulation/Distribution index continues to fall, this signifies that the volume is not following the price movements and the trend is relatively weak.

IMPLEMENTATION OF TECHNICAL INDICATORS IN METASTOCK

RECEIVING AND PROCESSING STOCK QUOTES

Historical prices of stocks for the purpose of technical analysis in this project were obtained from the Analyst's Dataserver version 3.4.3. It has a built-in Metastock format database of SGX stock data and worldwide stock indices and allows a direct download of end-of-day stock data from the Singapore Stock Exchange.

All the stock data is then downloaded for processing into Metastock version 10.0. This version of Metastock has quite a few built in indicators which can be automatically attached to the underlying security for analysis. There are various methods of doing this, through various Metastock Power Tools. Some of these are explained below:

Enhanced System Tester: The Enhanced System Tester is the main power tool that has been used for the purpose for this project since it allows the user to create and back-test trading systems. Once systems have been developed for the various technical indicators, they can be easily tested and compared on SGX stock data to better understand their performance. The entry and exit conditions are set according to the user's preferences. This power tool also the user to optimize the parameters of the systems to understand what works best for a security. This feature has been excessively used in this project to optimize the time periods of various indicators.

Indicator Builder: This is a built in feature of Metastock where the formulae and Metastock implementation of various indicators is already provided. The user can enter the parameters according to his/her preference and also create custom indicators. For this project, new custom indicators were built for some indicator systems while some have been used as they were provided by Metastock.

The Explorer: The Explorer is another powerful multipurpose tool provided by Metastock to enable the user to test a system and scan multiple securities at a time. It filters the securities according to the technical criteria of a system and shows the securities with buy and sell signals.

The Expert Advisor: The Expert Advisor provides the best way to automate a trading system by establishing rules, signals and alerts for the user. It also provides a commentary for easier and quicker understanding of the underlying security. The ultimate aim of this project is to develop an indicator system and implement it as an Expert Advisor.

In the course of this project, systems were developed for the indicators explained above using the Indicator Builder. These were then tested on historical SGX quotes and their results compared and analyzed.

Different systems have different advantages and disadvantages. While they may provide buy/sell signals through crossovers, histograms or divergences, it is generally advisable to employ only one rule at a time and complement it with another rule from some other indicator. For instance, while the MACD may give good results through the crossover rule, it may falter by the histogram principle. One should then try to implement a rule of maybe some other indicator to correct this. It is also essential that the different rules come from indicators which measure different aspects of the market - momentum, trend or volume.

IMPLEMENTING INDICATORS AS TECHNICAL SYSTEMS

Upon getting a clear understanding of the technical indicators, the indicator rules and signals are now to be implemented as systems. Metastock provides an easy way of doing this through its power tools - the Indicator Builder and the Enhanced System Tester.

INDICATOR BUILDER

The mathematical calculations of the indicators are converted into functions which are then used in the Indicator Builder. These rules have to be written in Metastock's formula language and can then be saved under that indicator name. Some examples of these functions are:

  • AD() - Calculates the value of the Accumulation / Distribution Index
  • CMF (Periods) - Calculates the value of Chaikin's Money Flow Index and takes time period as a user defined parameter.
  • Divergence (Data Array 1, Data Array 2, %Min Change) - Highlights if the values of data array 1 are diverging from the values of data array 2 over a period of one week. A min percentage change in the values of data array 1 as specified by the user is required to observe divergence.
  • MACD () - Calculates the value of the Moving Average Convergence Divergence over default time periods of 12,26 and 9.
  • MFI (Periods) - Calculates the value of the Money Flow Index by taking time period as a user defined parameter.
  • RSI(Data Array, Periods) - Calculates the value of the Relative Strength Index by taking the data array and time period as user defined inputs.
  • Stoch(%k Periods, %k Slowing) - Calculates the value of the Stochastic Index by taking %D and %K time periods as user defined inputs.
  • Typical () - Calculates the typical price of the day as required by the Money Flow Index indicator.

ENHANCED SYSTEM TESTER

Once all the functions required for writing indicator rules have been developed, they can be implemented as systems through the Enhanced System Tester. The Enhanced System Tester allows the user to put in buy/sell order conditions as well as sell short/buy to cover orders. Since the Expert Advisor mainly executes the buy/sell signals as entry points for any trade, this project shall focus mainly on them. However, users may set their own levels while using the system and Expert Advisor.

Another advantage of using the Enhanced System Tester is that one can put in stop levels defining maximum loss, breakeven conditions, maximum profit desired, trailing positions and inactivity minimum change. These stops have been left to the preference of the user. For the purpose of this project, all systems were tested with stops of 2%. This implies that if the price moves in the opposite direction as desired by the position taken by the trader, the system will execute the trade and remove that position.

TECHNICAL INDICATOR SYSTEMS

Now that the theoretical knowledge of the various indicators has been learnt, one can design the systems in Metastock using the Enhanced System Tester. The indicator principles are executed as strict rules and define all trading positions that the system takes on the underlying securities.

The technical indicator systems used for comparison in this project are:

  1. Moving Average Convergence Divergence (MACD)
  2. Stochastic Oscillator (SO)
  3. Relative Strength Index (RSI)
  4. Optimized Moving Average Convergence Divergence + Stochastic Oscillator
  5. Optimized Moving Average Convergence Divergence + Relative Strength Index
  6. Optimized Moving Average Convergence Divergence + Money Flow Index

All these technical indicator systems have been provided in the appendix at the end of this report. Please go through them for a better understanding of the following sections.

While the MACD, SO and RSI indicator systems are already provided with Metastock, the other three systems have been designed by the author. These new systems have optimized time parameters for the MACD indicator and this has been used as the basis of all systems designed by the author. In all these systems, the MACD crossover rules have been complemented with overbought/oversold rules of another indicator to provide more reliable signals.

The MACD indicator rules developed by the author mainly use the crossover signals of the MACD to highlight buy and sell conditions. After excessive testing and research, it was observed that the crossovers are the best signals provided by the MACD since it is inherently a trend following indicator. In comparison, the crossover signals provided by the Stochastic Oscillator (SO) and the Relative Strength Index (RSI) are weak. However, the SO and the RSI better highlight overbought/oversold conditions since they are both momentum measuring indicators.

Another method of obtaining buy/sell signals is divergence. Divergence signals are best provided by the Accumulation/Distribution indicator. This indicator provides reliable divergence signals as it measures both the price and volume of the trend, giving a better indication of momentum. Even though it is inherently a volume measuring indicator, it combines both price and volume effectively to be able to judge divergences more accurately.

However, in SGX stocks, where the trading has been restricted in the past year, it is very tough to obtain divergence signals through the Accumulation/Distribution indicator. For this purpose, the Accumulation/Distribution indicator rules have not been directly implemented in any technical system. Its rules are very strict and restrict the trading positions of the system to bare minimum. It is a very useful index though and one that cannot be ignored. For this purpose, the Expert Advisor, in its commentary box, displays if the underlying security displays divergence.

The author has written the indicator rules for the Accumulation/Distribution index and they can be observed in the commentary box of the Expert Advisor.

TESTING AND COMPARISON OF THE TECHNICAL INDICATOR SYSTEMS

STOCKS SELECTED FOR BACK-TESTING

For the purpose of testing and comparing the different technical indicator systems, some selected SGX stocks have been chosen. After a broad research of the many SGX stocks, the author has chosen those that show the maximum activity and best represent their respective sector. The SGX stock market lists the stocks by categorizing them into sectors, which are:

  • Commerce
  • Construction
  • Finance
  • Hotels and restaurants
  • Manufacturing
  • Multi-industry
  • Others - Services, Loans and Debentures, Agriculture, Transport etc.
  • Property
  • TSC (Transport, Storage and Post & Telecom)

The stocks selected for testing and comparison are:

  • Commerce: China Dairy Group Ltd., Olam International Ltd. and Pan United Corporation Ltd.
  • Construction: Hiap Seng Engineering Ltd., Shining Corporation Ltd. and Koh Brothers Group Ltd.
  • Finance: DBS Group Holdings Ltd., Hong Leong Finance Ltd. and Oversea Chinese Banking Corp.
  • Hotels and restaurants: ABR Holdings Ltd., Mandarin Oriental Intl. Ltd. and Food Junction Holdings Ltd.
  • Manufacturing: Sunningdale Tech Ltd., People's Food Holdings Ltd and Creative Technology Ltd.
  • Multi-industry: Singapore Tech Engineering Ltd., Hotel Properties Ltd. and Keppel Corporation Ltd.
  • Property: City Developments Ltd., Keppel Land Ltd. and Singapore Land Ltd.
  • TSC (Transport, Storage and Post & Telecom): SingTel, Starhub Ltd. and Singapore Airlines Ltd.

For all stocks and indicators, stock information for 500 periods prior to 17th February, 2010 has been included for testing. This represents the time period for which the author was working on this project.

Note: It has to be admitted that these stocks might not be the best choice for testing these indicators since the trading activity in the past year was slow and influenced heavily by factors outside the scope of this project. To better understand the performance of the indicator systems, a powerful fundamental analysis is very essential. The following few sections, however, present a pure technical approach to stock trading; how the technical rules fared against the market and which indicator worked the best. These systems have then been compared. For further analysis, a fundamental approach is required.

BACK-TESTING AND COMPARISON

The systems hence developed can now be tested on historical stock data. The systems were tested on all SGX sectors and their results compared and evaluated. The following section contains system performances as obtained from the Enhanced System Tester.

COMMERCE:

Systems tested: MACD, SO, RSI, Optimized MACD + SO, Optimized MACD + RSI, Optimized MACD + MFI.

Stocks: China Dairy Group Ltd., Olam International Ltd. and Pan United Corporation Ltd.

Testing:

Hence, in terms of average net profit, it can be seen that the best performer was the Optimized MACD + Stochastic Oscillator system. Since this is an optimized system, the best results were provided by the following time period parameters:

The above figure highlights that the Optimized MACD + SO system performs best when the time period parameters of the MACD are 14, 24 and 11.

Results: For the commerce sector, best results are produced by the Optimized MACD + SO system with the MACD time period parameter values for the faster moving average, slower moving average and exponential moving average (of the MACD - signal line) being 14, 24 and 11 respectively.

CONSTRUCTION:

Systems tested: MACD, SO, RSI, Optimized MACD + SO, Optimized MACD + RSI, Optimized MACD + MFI.

Stocks: Hiap Seng Engineering Ltd., Shining Corporation Ltd. and Koh Brothers Group Ltd.

Testing:

Hence, in terms of average net profit, it can be seen that the best performer was the Optimized MACD + Stochastic Oscillator system. Since this is an optimized system, the best results were provided by the following time period parameters:

The above figure highlights that the Optimized MACD + SO system performs best when the time period parameters of the MACD are 12, 24 and 7.

Results: For the construction sector, best results are produced by the Optimized MACD + SO system with the MACD time period parameter values for the faster moving average, slower moving average and exponential moving average (of the MACD - signal line) being 12, 24 and 7 respectively.

FINANCE:

Systems tested: MACD, SO, RSI, Optimized MACD + SO, Optimized MACD + RSI, Optimized MACD + MFI.

Stocks: DBS Group Holdings Ltd., Hong Leong Finance Ltd. and Oversea Chinese Banking Corp.

Testing:

Hence, in terms of average net profit, it can be seen that the best performer was the Optimized MACD + Stochastic Oscillator system. Since this is an optimized system, the best results were produced by the following time period parameters:

The above figure highlights that the Optimized MACD + SO system performs best when the time period parameters of the MACD are 10, 26 and 7.

Results: For the finance sector, best results are produced by the Optimized MACD + SO system with the MACD time period parameter values for the faster moving average, slower moving average and exponential moving average (of the MACD - signal line) being 10, 26 and 7 respectively.

HOTELS AND RESTAURANTS:

Systems tested: MACD, SO, RSI, Optimized MACD + SO, Optimized MACD + RSI, Optimized MACD + MFI.

Stocks: ABR Holdings Ltd., Mandarin Oriental Intl. Ltd. and Food Junction Holdings Ltd.

Testing:

Hence, in terms of average net profit, it can be seen that the best performer was the Optimized MACD + Relative Strength Index system. Since this is an optimized system, the best results were produced by the following time period parameters:

The above figure highlights that the Optimized MACD + RSI system performs best when the time period parameters of the MACD are 10, 24 and 7.

Results: For the hotels and restaurants sector, best results are produced by the Optimized MACD + RSI system with the MACD time period parameter values for the faster moving average, slower moving average and exponential moving average (of the MACD - signal line) being 10, 24 and 7 respectively.

MANUFACTURING:

Systems tested: MACD, SO, RSI, Optimized MACD + SO, Optimized MACD + RSI, Optimized MACD + MFI.

Stocks: Sunningdale Tech Ltd., People's Food Holdings Ltd and Creative Technology Ltd.

Testing:

Hence, in terms of average net profit, it can be seen that the best performer was the Optimized MACD + Relative Strength Index system. Since this is an optimized system, the best results were produced by the following time period parameters:

The above figure highlights that the Optimized MACD + RSI system performs best when the time period parameters of the MACD are 14, 28 and 11.

Results: For the manufacturing sector, best results are produced by the Optimized MACD + RSI system with the MACD time period parameter values for the faster moving average, slower moving average and exponential moving average (of the MACD - signal line) being 14, 28 and 11 respectively.

MULTI-INDUSTRY:

Systems tested: MACD, SO, RSI, Optimized MACD + SO, Optimized MACD + RSI, Optimized MACD + MFI.

Stocks: Singapore Tech Engineering Ltd., Hotel Properties Ltd. and Keppel Corporation Ltd.

Testing:


Hence, in terms of average net profit, it can be seen that the best performer was the Optimized MACD + Money Flow Index system. Since this is an optimized system, the best results were produced by the following time period parameters:

The above figure highlights that the Optimized MACD + MFI system performs best when the time period parameters of the MACD are 10, 24 and 7.

Results: For the multi-industry sector, best results are produced by the Optimized MACD + MFI system with the MACD time period parameter values for the faster moving average, slower moving average and exponential moving average (of the MACD - signal line) being 10, 24 and 7 respectively.

PROPERTY:

Systems tested: MACD, SO, RSI, Optimized MACD + SO, Optimized MACD + RSI, Optimized MACD + MFI.

Stocks: City Developments Ltd., Keppel Land Ltd. and Singapore Land Ltd.

Testing:

Hence, in terms of average net profit, it can be seen that the best performer was the Optimized MACD + Stochastic Oscillator system. Since this is an optimized system, the best results were produced by the following time period parameters:

The above figure highlights that the Optimized MACD + SO system performs best when the time period parameters of the MACD are 10, 26 and 7.

Results: For the property sector, best results are produced by the Optimized MACD + SO system with the MACD time period parameter values for the faster moving average, slower moving average and exponential moving average (of the MACD - signal line) being 10, 26 and 7 respectively.

TSC (Transport, Storage and Post & Telecom):

Systems tested: MACD, SO, RSI, Optimized MACD + SO, Optimized MACD + RSI, Optimized MACD + MFI.

Stocks: SingTel, Starhub Ltd. and Singapore Airlines Ltd.

Testing:

Hence, in terms of average net profit, it can be seen that the best performer was the Relative Strength Index system.

Results: For the TSC sector, best results are produced by the Relative Strength Index system.

INFERENCE AND ANALYSIS

In the above section, all the indicator systems have only been compared in terms of the average net profit. This gives us a pretty clear idea of how these indicators have performed in the past year. However, it is important to understand that these are just the technical rules and their actual success or failure would be very difficult to explain without a substantial fundamental analysis to back it up. For instance, even though a particular indicator might perform very well for a particular stock, its performance could prove to be dismal for another stock from the same sector. It might not even have any positions in any other stock since trading in the SGX market was very limited, especially in the previous year. While the MACD, SO and RSI systems of Metastock rely on only one condition to be true for undertaking a stock position, for all systems designed by the author, positions were only acquired if two separate conditions were fulfilled. Getting two separate conditions to be true was a rare occurrence. While this restricts the number of trades made by the system, it ensures greater accuracy and profit.

For a better understanding of the above system performances, an in-depth analysis of the various trades made by a system is required. The following section provides an analysis of the Optimized MACD + Stochastic Oscillator system's performance in the Construction and Finance sectors.

Construction:

It has already been observed that the Optimized MACD + Stochastic Oscillator, with MACD time period parameters of 12, 24 and 7, is the best indicator system for this sector. Let us now see in detail the various trade positions that it advised and undertook.

The figure above shows that the system produced gains for all stocks, while the maximum profit was garnered for Hiap Seng Engineering Ltd.

The system's performance on Hiap Seng Engineering Ltd. is now discussed in greater detail. The Result Details view explains the system performance and is provided in the appendix at the end of the report. The trade positions have been plotted on a chart and are examined below:

The arrows in the figure highlight the most successful trade position undertaken by this system. The system proposed the following trades:

Long:

Trade opening price: 0.1800 points ; Date: 16/09/2008

Trade closing price: 0.7000 points ; Date: 28/09/2009

Profit: 0.5200 points; 288.88%

Short:

Trade opening price: 0.7200 points ; Date: 28/09/2009

Trade closing price: 0.6300 points ; Date: 17/02/2010

Profit: 0.090 points; 12.5%

Hence, it can be seen that the indicator proposed highly successful trade positions and provided fair returns for stocks in the construction sector of SGX. This once again emphasizes the relevance of technical analysis in the stock market today.

Finance:

It has already been observed that the Optimized MACD + Stochastic Oscillator, with MACD time period parameters of 10, 26 and 7, is the best indicator system for this sector. Let us now see in detail the various trade positions it proposed and undertook.

The system's performance on DBS Group Holdings Ltd. is now discussed in greater detail. The Result Details view is provided in the appendix at the end of the report. The trade positions plotted on a chart are shown below:

The arrows in the figure highlight the most successful trade position undertaken by this system. The system proposed the following trades:

Long:

Trade opening price: 7.8960 points ; Date: 20/11/2008

Trade closing price: 14.5600 points ; Date: 21/12/2009

Profit: 6.6640 points; 84.4%

Short:

Trade opening price: 14.800 points ; Date: 21/12/2009

Trade closing price: 14.200 points ; Date: 17/02/2010

Profit: 0.600 points; 4.05%

Here, once again we have seen that the indicator system has come up with highly positive results. This makes the Optimized MACD + Stochastic Oscillator the most efficient technical indicator system.

As explained earlier, the Optimized MACD + Stochastic Oscillator uses crossover confirmations from the MACD but implements a trade only when the underlying security is in an overbought or oversold region, as per the Stochastic Oscillator indicator. Getting both conditions to come true simultaneously is rare due to which this system makes so few trades. However, more often than not, these trades come up with highly successful results, as we have seen above.

Similar analysis of all other trades can be carried out using the systems provided in the appendix at the end of the report and carrying out the procedure as explained above. The author has only explained the most profitable results; there are instances when the same system would fail and it is very interesting to study the market conditions then.

THE EXPERT ADVISOR

Now that all the technical indicator systems have been tested and compared, their rules can be implemented as an Expert Advisor. Experts can be attached to the chart of the underlying security. This applies the system rules to the data in the chart and the user can observe all sorts of trends, symbols and highlights in the movement of the stock. The Expert also generates alerts for the latest data. Signals, however, appear for all the data, dating back to the first period of observation. An Expert Advisor in Metastock has the following features:

Symbols: buy and sell signals are pointed out on the chart with arrows.

Highlights: the price bars are colored green while in long positions and red while in short positions.

Alerts: a dialog will pop up notifying the user of any buy and sell signals.

Commentary: a description of the buy and sell signals.

The figure above shows an Expert Advisor attached to the chart of DBS Group Holdings Ltd. The bottom window displays volume information; the middle window displays chart information with buy/sell signals from the attached Expert Advisor and the upper window displays the Expert Advisor equity. Overbought/oversold conditions are highlighted by the red and green colored ribbon at the bottom of the window. The Expert commentary can also be displayed along with the chart information and it plays an important role in describing the buy/sell signals and market conditions. The Metastock Expert Advisor allows the user to set these signals, highlights and alerts to their preference (color, symbol, sounds).

For this project, six systems were selected for testing and comparison and it was observed that for different stock sectors, different systems produce the best results. For this purpose, the author has designed different Expert Advisors for use in different sectors. Since the user designed systems had optimized time period parameters, the best parameters have been selected for implementation in the Expert Advisor.

Six Expert Advisors have been designed by the author:

  1. Optimized MACD + Stochastic Oscillator I (time period parameters - 14, 24, 11)
  2. Optimized MACD + Stochastic Oscillator II (time period parameters - 12, 24, 7)
  3. Optimized MACD + Stochastic Oscillator III (time period parameters - 10, 26, 7)
  4. Optimized MACD + Relative Strength Index I (time period parameters - 10, 24, 7)
  5. Optimized MACD + Relative Strength Index II (time period parameters - 14, 28, 11)
  6. Optimized MACD + Money Flow Index (time period parameters - 10, 24, 7)

All Expert Advisors are provided in the appendix at the end of this report.

As explained earlier, the basis of all systems designed by the author is the Moving Average Convergence Divergence (MACD) indicator. Hence, for all Expert Advisors, the trends are highlighted by the basic MACD definition of crossovers of the MACD with its signal line. Overbought/oversold conditions are highlighted either by the Stochastic Oscillator, Relative Strength Index or Money Flow Index indicators. The buy/sell signals are generated only when both conditions, i.e. a MACD crossover in an overbought/oversold region, are satisfied. Since the Accumulation/Distribution indicator was not implemented directly in any system, it is written into the commentary and indicates if the underlying security is undergoing any divergence. This indicator has mainly been used to indicate if the buy/sell signal as generated by the Expert Advisor can be confirmed by the Accumulation/Distribution indicator. Since the MACD is a trend following indicator, and the Stochastic Oscillator and Relative Strength Index are momentum following indicators, the volume following Accumulation/Distribution index should provide a reliable confirmation of the signals generated by the Advisor.

THE EXPLORER

The Explorer is another important power tool provided by Metastock. It allows the user to set up signal explorations to look for stocks that have generated buy/sell signals for a system. All the systems that have been implemented as Expert Advisors by the author have been tested on the Explorer as well.

The Explorer allows the user to select stocks for the Expert Advisor that is being used. It was seen in the previous sections that different indicator systems perform differently for the various SGX stock sectors. The Explorer is hence a very useful tool for selecting the stocks relevant to a particular Expert Advisor system.

Performance explorations were run by the author for all Expert Advisor systems on all SGX stocks, with the previous 1000 time periods being included for analysis. Filters were set for crossovers and overbought and oversold conditions according to the technical system being used. The list of stocks, which showed the best results, and the stocks, that were rejected for use by the system rules, were discovered and saved by the author in the Explorer system.

The list of SGX stocks that work best with an Expert Advisor system, and those that are rejected, are provided in the appendix at the end of the report.

CONCLUSION AND FURTHER WORK

CONCLUSION

From this project the author concludes that trading using technical indicators can prove to be very profitable but also complicated at times. Even though the technical indicator rules work in certain situations, they fail in most. The different indicators react differently to different market conditions (for instance volatility) and their performance differs for different stock sectors.

The author discovered, through the technical analysis of various securities in Metastock, that trading using only technical indicators results in losses in the long term. Since technical analysts base their expectations more on past earnings or the previous track record, they tend to ignore the way the market works. It is generally seen that future stock prices are strongly influenced by investor expectations and market movements of the past. Because of this, analysts claim that past prices play a major influence on future prices. [11] They also point to research in the field of behavioral finance specifying that people do not always make rational decisions. Analysts believe that it is this irrational human behavior that influences stock prices, and that this behavior leads to predictable outcomes. [12]

That being said, the markets in the previous year were in no way predictable and any sort of analysis is bound to produce some very interesting results. As explained earlier, for a complete understanding of successful trading in the stock market, technical analysis should be well complemented with sound fundamental analysis of the market conditions. This is even more essential for the time period of analysis of this report since the markets were more influenced by factors external to the scope of this project. The technical indicator rules imposed on the market were still the same, and their failure in these conditions in no way reflects their performance as a whole. The same systems that are failing now might come up with outstanding results as the market variables change. This is the beauty of stock trading - it is unpredictable and there is a lot to learn.

The author thoroughly enjoyed working on this project as it gave him an opportunity to explore the financial world with a technical approach. This has increased his understanding of the global markets and the effort that goes into successful stock trading. The author hopes to repeat this research for the next few years and follow up on the results to gain an all round understanding of the technical indicators and systems created.

FURTHER WORK

The scope of technical analysis of financial instruments is almost endless. While this project was developed in strict adherence to the author's initial objectives, various modifications to the current objectives may lead to various other projects. The following are some areas which may be looked into in further projects:

  • Intraday Trading: Technical analysis should next be applied to intraday trading whereby the technical indicator systems would generate various buy/sell signals in one day. The challenge here would be to identify the indicator time periods for assessing time periods.
  • More technical Indicators: More Technical Indicators may be introduced to check their performance and profits. Some of these indicators are Bollinger Bands, Elliot Wave Analysis etc.
  • Fundamental Analysis: No trading system can be effective in the real market if not coupled with fundamental interpretations of the stock conditions. The project would be more comprehensive by including a fundamental analysis of the market conditions too. This would make the system more secure and ready to be reliably used in the market.
  • More Financial Instruments and Markets: Lastly, the indicators could be tested for trading in other financial instruments such as FX, Futures, Options, Bonds, etc.

REFERENCES

  1. http://www.technicalanalysis.org.uk/quotes.html
  2. http://www.incrediblecharts.com/technical/support_resistance.php
  3. http://www.incrediblecharts.com/indicators/moving_average.php
  4. http://www.automatedtrader.net/glossary/Head_and_Shoulders
  5. http://en.wikipedia.org/wiki/Triangle_(technical_analysis)
  6. http://en.wikipedia.org/wiki/Wedge_pattern
  7. http://www.swissquote.ch/static/help/tools/charts/macd_e.html
  8. http://www.tradingmarkets.com/.site/stocks/how_to/articles/How-to-Use-the-Stochastic-Oscillator-to-Trade-in-V-80036.cfm?lid=right-home-popular-2
  9. http://www.textbiz.org/moneyflow.html
  10. http://www.tradingstocks.net/html/relative_strength.html
  11. http://www.jimwyckoff.com/articles/traders/cardwell.html

Please be aware that the free essay that you were just reading was not written by us. This essay, and all of the others available to view on the website, were provided to us by students in exchange for services that we offer. This relationship helps our students to get an even better deal while also contributing to the biggest free essay resource in the UK!