HOW TO MAKE A MODEL OF 3 INDICATORS

Someone may want to trade using daily stock data, for finding the daily trends in stocks. This opposes the view of longer term traders/ investors who may choose weekly, or monthly data for the work.

Indeed, the advice that goes around in the lobbies of traders/ investors is that it may be better for someone to be an investor rather than a trader and also use longer term data to catch the long term trends of the stocks. For a trader who tries to trade using daily data, the randomness of daily data starts to become a problem. There is a difference in the way of trading stocks for short- term traders and longer- term investors in that that usually the short- term data can move erratically up, or down, i.e. that in a short amount of time, fortunes can be made or lost due to the speed that the stock may be moving. On the other hand, longer- term investors are happy with catching the longer term trends which move less randomly.

First of all, in order to trade short- term data, i.e. daily prices a good model may be needed. A good model in this respect is 3 indicators used together Plus the addition on protective stops since we are dealing with short- term data and can move erratically up or down.

The first indicator can be an oscillator, or ROC. This is also called speed, or velocity oscillator. This indicator will need optimization using backtesting software to find the best fit for it according to the stock being traded and the older data of the stock.

The second indicator can be a momentum indicator. Momentum oscillator is any convergence/ divergence indicator. Again this can be optimized using backtesting software.

The third indicator can be a bands indicator, i.e. either percentage bands above/ below a moving average, or simply the use of Bollinger bands. This indicator again will need to be backtested to find the best optimization curve to fit the stock data.

And lastly, as I mentioned, a type of protective stops can be placed to exit the market when the data become too volatile, or to protect the gains when the gains are running quickly.

This is a good plan for trading short- term stock market data (daily values of stocks). The entire model can be backtested as well with a more sophisticated backtesting data that can backtest 3, or 4 indicators together.

Furthermore, the need to run backtests again will be more meaningful if done once every month for example, so that the new data of the stock are incorporated into the system, while the older historical data are discarded.

---------

[Because I dont get paid for writing in my blogs, I would request perhaps any readers who may find the topics discussed here as interesting, then they may want to contribute some money so I have available resources to write more topics. If you can submit a small payment for example to my paypal account email chrisk144@gmail.com then I can continue writing my blogs without delay, or hesitations because of my busy time/ schedule, or lack of resources.

Thank you very much]

ARM' S EASE OF MOVEMENT

Richard Arms is the creator of this indicator. His intent was to quantize the market prices together with volume prices with n period price/ volume etc.

The mathematical formula is as follows:\

EMV = [ [(H + L) / 2] - [(Hn + Ln) / 2] ] / [V / (H - L)]

where n is the n- day period.

The formula for Arm' s reach of movement is rather complex. The results, when run by backtesting of historical data were erratic to say the least. The purchases and sales of the stock were frequent which means that the indicator although was buying/ selling the stock, yet profits were not up to par to warrant to call this indicator useful.

Another way to follow this kind of indicator is to apply the m- day moving average to the EMV value above so we have the Moving Average of EMV if we want to smooth the data a bit. Because EMV depends on only one value in the past, it may have the same pittfalls as other indicators who do the same, i.e. rely only on one value from the past. Thus, applying an MA to smooth the data seems like a good idea.

The results were not very ecouraging, for values below m=10 the indicator was erratic, and for values greater than m=40, again the behavior of the indicator showed some randomness and losses.

Best values was the 14-day Moving average of the n=1 day EMV.  Values higher than n=1 were not calculated by the backtesting computer software which means that there might be room for improving this indicator in the future. Someone may want to study again this indicator and run backtesting for all values of n as well, like they were done for m.

In overall, the profitability graph of the backtesting tests were slightly bell- shaped showing behavior that both values before and after m=14 were erratic. Therefore, it may not show any reliable results.

STOCHASTICS

The Stochastic indicator is a popular indicator used to predict stock trends, or overbought/ oversold states. The Stochastics indicator is similar to the ROC (Rate of Change) indicator and also has the negative features that ROC has as well. Because both indicators rely only on one data from historical charts, then it may be prone to instability and noise.

The Stochastics indicator is calculated by subtracting the today close price of the stock by its n- day Low value and subtracting the n- day High value of the stock from the n- day low value and dividing the two, i.e. forming the ratio, or rate of change/ momentum of the stock. The value is then converted to percentage by multiplying with 100. N is the number of days in the past.

There are 2 more values to the Stochastic indicator, it is the 3-day moving average of the above value which is called SK and again the 3 day moving average of SK forms the SD value.

Signals of change of position in stocks are complicated with this indicator because the trader needs to see both convergence, or divergence of the stock price with the Stochastic indicator as well as when SK crosses SD and also when the signals are given when the Stochastic indicator is at the 10%-15% range, or the 85% to 90% range.

Actually from backtesting results this indicator proved to be lossy, i.e. it lost the invested capital rather than preserving it, or increasing it. Even when all rules above were tested still the indicator did not provide any good results.

Another practical rule of Stochastics is that SD to be above 80% for a sale and below 20% for purchasing the stock. Even these rules though provided consistent losses.

Many people who studied Stochastics mentioned that its validity in predicting stock changes in price, may be overstated, i.e. this is not a useful indicator. Therefore, it may be better that traders who may use it to be careful, or that they not use it at all. 

NEW HIGHS, NEW LOWS

New Highs- New Lows is a simple indicator that tracks the highs and lows of a stock depending on a number of days. In other words, if the stock has formed a new high, or new low in the last n days, then it may be considered bullish, or bearish respectively.

A number for setting n is about 52 days, in other words, if there is a new high, or new low that it occurred today from the last 52 values, then it is a reason of buying and selling the stock.

Another way to do this indicator, is to first form the MA of price data, let' s say the 10 day MA and then we use the MA to see if it has done new highs, or new lows with respect to the data within the last 52 days. This way some inconsistencies and randomness that may exist in daily prices, it is smoothed out by the MA.

What we can deduce is that there is correlation between new highs that the market is bullish and that new lows respectively correspond to slightly bearish market. Backtesting this indicator, someone can create the denominations of Very Bearish, Bearish, Slightly bearish, Slightly bullish, Bullish, very Bullish with respect to the values of this indicator, i.e. the gain, or loss of the backtest. The results though may not be consistent enough to time the market with purchases, or sales of stocks. Most consistent results were those of Slightly bearish and Slightly Bullish, i.e. this indicator was able to keep track in the best way the Slightly Bearish and Slightly Bullish associations.

In general, this indicator is complicated and uses statistical techniques to compare the various values of gain, or loss that it had. Novice traders may want to view other more straight forward indicators rather than depend on this indicator for buying and selling. It can only be used as a sentiment indicator, rather than a market- timing indicator, i.e. when to buy and sell.

The conclusion of this indicator is that very high new highs are considered slightly bullish, while very low new lows can be considered as slightly bearish. 

EXPONENTIAL MOVING AVERAGE (EMA)

The exponential moving average is a popular indicator. It is calculated in the same way as the Moving Average is calculated but with a weighing factor which is the exponential component. Thus, the result is data weighted by time, i.e. the recent data affect the indicator value more than the past data. It is the same kind of arithmetic mean formula but with a weight that becomes less and less when the data are older.

There is only one rule to use the EMA for trading and it is either the direction of the indicator, i.e. whether it is going up, or down, or the crossover of the price with the indicator.

All kinds of Moving Averages may have one pitfall actually that they take take into account stale data of the past, thus some care should be taken when dealing with Moving Averages that the arithmetic series is not too long in the past so as to add also unimportant data. Therefore, the n- day moving average needs to be short in sequence, rather than choosing the n value to be high. A lower n value should bring better results. It has been found that a value for n that is not to short, but also not too long may bring the best results and traders can say that a value of n around about 30 to 50 days, which makes the average good period to be about 42 days. Therefore, I will talk here for the 42-day EMA.

The 42 day EMA produces a large number of transactions compared with other indicators, but net profits are higher than other indicators, like the Simple Moving Average (MA). Therefore, it may have some advantage but because of the large number of transactions, profits per transaction was a bit low. Therefore, this may need to be traded for a longer term in order to make substantial profits, when compared with other indicators. For the short- term trader the EMA may not be very useful, while for someone who is trading stocks for the longer term, then the EMA will make sense to be used so that profits are taken from longer period.

ON BALANCE VOLUME (OBV)

On Balance Volume is an indicator which adds up, or subtracts the volume of the day to create a cumulative Volume indicator. When the price of the stock for that day has gone up, the V takes up a positive sign and it is added to the cumulative volume from the previous days. When the price of the stock has fallen for the day, then the volume takes a negative sign and it is subtracted from the cumulative value. Thus, OBV is a volume indicator that is dependent on the price of the stock as well.
OBV resembles the stock prices themselves since also stock prices are added/ subtracted every day thus we get the effect of stock price going up, or down.
With the same rationality, OBV goes up, or down as well.

The OBV is used in various way, one way is by identifying the stock trend and another way by identifying the strength of the stock, i.e. whether volume increases when price increases, or decreases when stock price decreases. Divergence from this also is useful, thus OBV is used to show an overbought, or oversold state at the current price of the stock.

Because daily volume may be thought to be random in the same way as daily stock price data are are also thought to be random, then a Moving Average of the OBV is taken to smooth out the curve. Therefore, the Moving Average of the OBV is usually the effective indicator to look at.

Sometimes, the Moving Average of the OBV is compared to the raw OBV and decisions for buying/ selling may be made this way, when the OBV crosses its own n- day OBV.

It can be said that this is a good Volume indicator to smooth out any daily volume data which cannot be trusted fully. Also, because this indicator takes in consideration also the stock price change, then it becomes useful in this respect, i.e. is dependent on stock price direction as well.

Weekly data can be taken as well. A popular value for n for the MA OBV is 9 days, therefore the 9 day MA of OBV is calculated this way.

OVERBOUGHT/ OVERSOLD INDICATORS

Overbought/ Oversold indicators show the current trend of the stock cannot be sustained for further. That means that they signal either excessive buying of the stock, or excessive selling of the stock. Thus, they are indicators that signal reversal of the current trend of the stock.

The indicators can be helpful, but there is ongoing discussion as to whether all traders/ investors may find these characteristics useful for buying/ selling the stock. The reason behind the questioning of the overbought/ oversold indicators is that overbought, or oversold states, may not show that there is a reversal of the trend being imminent, but that there may be stronger conditions for the stock to advance even further in the current direction. Therefore, there is a clash of opinions whether these indicators are useful, or not.

Advanced stock traders, may warn that for inexperienced traders, they should shy away from consulting such indicators because of the difficulty in determining if they signal reversal of the current trend, or a push further, i.e. to higher values in the case of overbought, or lower values in the case of oversold.

AS a conclusion, we might say that it takes experience to determine what the stock may be doing next and that novice traders/ investors should probably not work with these indicators, or at least confirm the direction of the stock with more professional traders, so that they have at least a second opinion on the movement of the stock.

Actually overbought/ oversold states can be formed by many popular indicators like the ROC (Rate of Change). The ROC determines the speed of the stock, thus if the stock shows high speed in the direction of the current trend, may mean obvioulsy overbought traits, since this is what drives the stock further, or increases its speed. The same on the down trend, if the stock is showing a high speed of losing value, it may not necessarily mean that it will change direction shortly but may mean that the stock is going to go even further lower.

Therefore, as a conclusion we may say, that following these indicators without proper experience, may result in cutting winnings short and may allow the losses to run further.Therefore, in overall, it is imperative, or suggested/ recommended that such indicators, not be taken at face value, but to be used only as filters in filtering out any decisions that are taken by more appropriate indicators.

TRADERS VS INVESTORS

In continuation of the previous article, when we talk about investing, we mean to hold the position of a stock for an extended amount of time. In this approach we may mean weekly and monthly data from stocks.

Now there is also the shorter run adoption of viewing the daily data as also described in the previous blog post. This is the difference between a trader and an investor, i.e. the notion that investors will hold the stock for longer amount of time as well as view the fundamental analysis of the stock rather than the technical analysis.

The reason for this distinction is that daily data are considered to be more random than weekly, or monthly data. On the other hand, we cannot ignore the fact that even daily data may have trends, so this is what traders try to do, i.e. follow the more volatile daily data. For very volatile stocks, because of the greater movement they do in daily prices, a lot of money may be made - or lost - in a short amount of time, i.e. days. This is the logic behind daily data compared with weekly and monthly data.

The investors who deal with the longer run of the stock, may argue that traders may be seen as... "taking blind leaps of faith" when seeing daily data, or that they are dealing with a... "black box" i.e. something that may not be showing its internal characteristics.In other words, the stock may depict random behavior when seen by its daily data.

Additionally, when dealing with daily data, traders believe that protective stops are useful as well, something that longer term investors may not be using due to the different nature of their investing. Thus, traders argue in return that with good protective stops placed at strategic intervals and prices, trading becomes a safe means of making money.

DEMAND INDEX

The Demand Index is a lesser known index. It tries to follow price and volume values of a stock. It is a momentum indicator of price and volume, The mathematical derivation of Demand Index is complicated, thus it will not be covered here, but later on at another post.

Useful values for n, i.e. the number of days for the Demand index to be calculated, seem to be in the 40- 49 days. These may give the highest profit when backtested. Backtesting is a computerized procedure to optimize indicator values and it will be discussed in a subsequent article.

The Demand Index involves the Moving Average values as well, thus usually a 40 day, or 49 day average can be used to calculate the Demand Index.

Demand index is a proprietary indicator, copywrited to its creator, James Sibbet.

During backtesting, it did provide some consistent results but not as much as other popular indicators.

The indicator showed some randomness with various values when backtested which were alarming i.e. dangerous to deal with. In other words, the drawdown was substantial. Therefore it needs to be traded with caution because it may reach to lose a significant amount of money before actually making any profit.

In general, Demand Index needs further studying and understanding to be used for trading. In a subsequent article we will deal with the calculation of Demand Index and some examples of how it performed. The way it is traded is by buying long and selling short when the signals are triggered, thus it makes money even when the stock is falling.

In overall and for time being, it may be ignored, as it is termed as one of the more complex indicators out there. Professional traders may want to study further the Demand Index indicator rather than casual traders to involve with it.

CPRs - CHART PATTERN RECOGNITION

Chart Pattern Recognition is a discipline on its own, but it falls under the broader term of Technical Analysis because CPR is in some respect a technical analysis system.

There are many CPRs that exist. Here in subsequent articles, I will be presenting a few of them.

CPR techniques signal reversal of the stock, So they are used for this reason. All other indicators of technical analysis are used for following trends, on the other hand CPR is used mostly for trend reversal.

Just because CPR techniques may be difficult to follow, there may be an inherent risk in using them for predicting stock cycles, so alongside CPR indicators, usually protective stops are used as well.

Protective stops are automated means to exit, i.e. sell a stock if some reversal of the trend appeared before the indicators themselves signaled a stock reversal. We will talk about protective stops in other article/ s.

One way to describe CPR techniques is that they may be more volatile, i.e. more erroneous than traditional technical analysis indicators. Mostly they are used by "swing" traders rather than technical analysts, i.e. the stock is kept for the very short run.

There is the common belief that by themselves CPR techniques may present with errors, rather than follow the market directly and concisely, and also without protective stops, CPRs may not be worthwhile. This is the view that technical analysts argue.

On the other hand, CPR indicators may be fun to utilize and observed them as they require only some knowledge of trading stocks. Pattern observing in the market is interesting to follow and with protective stops placed at strategic points they can be a useful tool for either novice, or advanced technical analysts. The reason that CPRs may appeal to novice traders is because they are easy to spot, or work with and require little mathematics, or no mathematics at all. But, novice investors may not have the maturity, or ability yet to discern correct stock patterns, thus they may be liable to losses. On the other hand, advanced technical analysts may also want to work with pattern recognition so that they double- check on their results from mainstream technical analysis.

Thus, as a conclusion, CPRs may be a good tool for all kinds of investors. One additional aspect of CPRs when applied to stocks, is that they tend to be outside of the market for the longest run, and they only briefly create buy signals to enter the market, i.e. to purchase the stock. Because as I wsaid previously, swing traders may benefit more from CPRs, the reason is that buy and sell signals for CPRs are brief and thus for most of the time, they signal the trader to be outside the stock market, i.e. to hold no position. That is why those who try to utilize CPRs may change to become swing traders, i.e. those who may change position of the stock very frequently.

AS a conclusion, CPR techniques may bring good results, especially if they are used as a system of various CPRs in unison, i.e. various CPR techniques to be used to verify one another. Caution is needed though, i.e. to knowing also when and where to put protective stops, For protective stocks I will write another time in a different article.

DAILY VS MONTHLY STOCK DATA

When checking stock data, it is best to look for the frequency of the observations as well, i.e. whether the data are daily, weekly, or monthly.
This depends on the time frame that the person wants to concentrate on, i.e. for how long he will be intenting to trade the stock and whether he will be keeping the stock, i.e. for few days for profit taking from stock' s daily ups and downs, or whether he eants to invest for the longer term.

If trading with daily data, then the person may be thinking more towards speculation, i.e. taking advantage of the stock' s daily ups & downs, rather than genuinely waiting for the stock to appreciate, thus giving profit in the longer run.

So, if a person is going to be keeping the stock for only a few days, the it may be useful to be checking daily stock data. If the person though may be willing to buy and keep the position for longer term, i.e. months, or even for a year, or more, then daily data may not be appropriate for him. For this strategy, i.e. the owning of the stock for a longer time, then the person may be more an investor, rather than a trader. Therefore, looking at a different frequency of stock data may be appropriate, i.e. the weekly values of the stock, or even the monthly.

Checking stocks is different than checking market indicators like the S&P 500, or Nasdaq, etc. If someone is to buy indices and be willing to keep the positions for years, let' s say, then the stock data he will be watching may not be daily data as they may be too erratic anyway, even for short- term traders who may be hoping to profit from daily values and ups & downs. Data like weekly, or monthly stock values may have more relevance and be less erratic than daily data, thus, may present a better view and better approach to investing, which means the long term keeping of the position of the stock, i.e. in general buying long as it is termed, i.e. buying for the long term.

Therefore, this is the use of monthly data values, i.e. they may be more worthwhile values to the investor rather than the trader. The monthly data may help also filter out any random prices in the stock and help to give a better view, of the longer term that the stock may be trending. The trend viewed will be a longer term view, as well as the absence of daily erratic data.

This approach, i.e. the view of longer term values for stocks, i.e. monthly values, also helps to see, or predict any recessions that may happen in the future, i.e. the person with monthly values, may be able to seeing the actual bull, or bear trends, i.e. whether there is booming market, or whether the may be a recession coming. Knowing any knowledge about bull & bear markets can make a person better fit for trading such cycles, which can be very important.

Thus, as a conclusion, we can say that the trader, or investor, may also want to see the long term ups & downs of the stock i.e. its monthly data alongside the daily data. Although daily data may be good for the short- term trader, the longer term data may be showing a better view of whether we will be living through a market boom, or market recession. Then, for example, if a recession is forecasted, even the short-term trader may want to avoid doing an purchases, if the stock may be having a longer term downward trend. In making money during a recession can be more difficult than making money during a boom so, more care may be needed, and perhaps the person to avoid completely the stock market and look to invest in mutual funds for example, or something else, like currency, or gold, etc.

BOLLINGER BANDS

Bollinger Bands takes its name from the creator of the bands, John Bollinger.

Bollinger Bands are a statistical indicator, drawing its use from Mathematics branch called Statistics. Using statistical techniques to forecast the financial markets is a popular way to build models for analysis. Being statistical indicator the Mathematics of Statistics apply and use of this indicator is derived from statistical techniques which are applicable not only for stocks but also for other applications. Therefore, it is a diverse way to follow, or predict the market.

Knowledge of statistics may make this indicator easier to follow, but just for using this indicator, no prior knowledge of Mathematics, or statistics is needed. For someone who wants to have more insight into how this indicator tracks the stock market, or why it is set up the way it is, then they can refer to Statistical analysis and mathematical books.

Bollinger Bands are drawn on the stock chart itself and set up living of how high, or low the stock may go. That is why they are called bands, because they form bands higher and lower from the stock price.

The good thing about Bollinger bands is that they use another indicator to plot the values, namely the Moving average. So a moving averate, let' s say the 30 day moving average is used first and then the Bollinger bands are drawn above and below the MA by some statistical values.

Namely, the Bollinger Bands use the Standard Deviation from Statistics theory, to draw the bands above and below the Moving Average.

It is interesting that standard deviation plotted against the stock chart itself, may be adjusted to include 90% of the data within the bands and 10% only to lie outside the bands. This is also the way the chart is predicted, i.e. if the stock appears to be outside the Bollinger bands, i.e. either the High, or low Bollinger band then it may mean reversal of the stock, or may mean that the stock may have increased a lot, or it has high speed moving in that direction.

The recommended setting for Bollinger Bands, is two Standard Deviations above and below the Moving Average.

Also, the bands tend to narrow if the stock is not showing trends, and then they may become broader when the stock show strong trends. So, a popular way to use the Bollinger Bands, is also by buying if the Bollinger Band is thin, or to sell when the Bollinger band becomes very wide.

In overall, Bollinger Bands are good statistical means of predicting the stock market.

RATE OF CHANGE (ROC)

Rate of change is a technical market indicator which is of interest to technical analysts. Mathematically it is the division of the price today with the price n days ago. Particularly we may take n to be 30 days.

Another way of calculating rate of change is by sinply findind the change in price over/ divided by the n, i.e. the number of days. For example, if the price today is $100 and the price n days ago was $80, then the indicator can be calculated as follows, i.e. (100-80) / n = 20 / n and if n is taken to be 30 days, then 20 / 30 = 0.66. If someone wants it as a percentage, then they can multiply by 100 which it will give 66%.

Trading rules for this indicator may need to see the stock how it is doing and find the trends and lowest/ highest values and note the ROC value. Then from this it may be fairly simple to find the turning points of the stock and devising a trading rule, i.e. trying to follow the natural cycles of the stock by noting down the values of the ROC when the stock is turning to go up, or down. Then we can use the values of ROC as indicators for buying/ selling, i.e. changing our positions of the stock.

ROC may be particularly interesting because it shows the velocity of the stock, i.e. how fast is changing for that particular number of days, i.e. how fast it is going up, or down. By taking the maximum values of this indicator, i.e. finding the fastest time that it goes up, or down, then we can build an indicator to buy, for example, when the stock is going up fast, or sell when the stock is falling fast.

The ROC may bring some erratic behavior to the stock, as you can see from the calculation above, it is only relying on one value from the past to calculate the ROC. Since daily stock prices may be erratic anyway, then using one single stock price value from the past to create an indicator, it may just bring some error to the calculations and exactly when the stock is turning to go up, or down.

Overall, it can be said that it is a good indicator to add to the arsenal of judging the stock and its ups and downs. This indicator my go hand in hand with the Moving Average indicator for example and help to validate the Moving Average indicator as well, before placing a trade.

In overall, it is a useful indicator.