Experiments in Technical Analysis

[SGM]

THANK YOU FOR THE PDF....I FOUND VERY INTERESTING & THIS WILL HELP ME\ EVERYONE A LOT...

Hello Casoni

The concept of giving utmost importance to Trade Management(position sizing/money management/defined stops) while using a very basic entry/exit criteria is indeed very intresting. It seems that consistently (& mechanically) following the system is the “Holy Grail” of trading success.

Regards
Sanjay

 

[karthikmarar]

Hi SGM,

This Thread is becoming more interesting with you, saji, marcus providing valuable information.

It very true that consistently following a simple trading system with good money management and disciple would be best way to trade.

Tough not related to pure technical Analysis, I would like to elaborate on this important topic a bit.

To give you an example why Money Management is important for trading. Let us look at how limiting your loss affect a trading system.

The expectancy of a trade is defined as

Expectancy = (Probability of win * average Money made)- (Probability of loss* Average money lost)

Suppose if we consider the probability of winning and loosing is the same , ie. 50%.

So the formula would be E= (0,5* money made) – (0.5* money lost)

Assume that you will implement a strict stop loss and will restrict the loss to a maximum of 10% of your money invested. But at the same time you will let your profits run and work on a risk to reward ratio of minimum 1:2, meaning that you will make at least 20% on winning trades.

So what will be the expectancy of this system?

E= (0.5*0.2)-(0.5*0.1) = 0.1 – 0.05 = 0.05

What it means is that you would make 5 paise for every rupee invested. Suppose you restrict you loss to 5% then you would make 7.5 paise for every rupee invested.

Interesting isn’t it? Even with the probability of win and lose being the same you would make more money if you strictly implement stoploss. This clearly brings out the importance of stoploss, which is a key factor in Money management.


It would be interesting to see if we have strict stop loss of 5% and also a system, which gives a winning probability of 60% (6:4), then you would make 10 paise for every rupee invested. At the same time if you make 30% in winning trades instead of 20%, you would make 16 paise per rupee invested. So it is important to make the maximum in winning trades (where exit philosophy comes into picture) and restrict your losses with strict stop loss. Play with different figures in the expectancy equation and you will realize how important it is to cut your losses per trade.

Warm regards

karthik

 

[saji oommen]

Hello,


To supplement more on karthiks explanation on stoploss and moneymanagement i have attached a wordfile . It should be beneficial.

Regards

Saji

 

[aad]

once the price moves up what can be the various methods to retain the profits i.e not to exit after loosing most of the profits gained. I am not trying to catch the entire move but yes how can be retain most of it.

Dear Rahulg77,

As pointed out by Karthik in his answer to your question, trailing stop-loss is one of the ways to protect your profits (as well as capital), once prices start going up. I am giving below a formula that I use in MetaStock (as an indicator). I just move my stoploss figure to the one shown by this indicator. I call it SafeZone

I have derived it by reading "Come into my trading room by Alexander Elder", with some modifications. Expect everyone to comment on its usefulness as trailing stoploss.

-------
multiFactor:=Input("Enter factor of agressiveness",0.5,2,1.5);
stopLoss:=If(L>=Ref(L,-1),Ref(L,-1)+multiFactor*(L-Ref(L,-1)),Ref(L,-1));
Mov(stopLoss,9,E);
-------

Here, if you use higher factor of agressiveness, say 2, you will be able to lock-in max. profits but a little higher volatility during the day may square-off your long position. Generally, I have observed that a factor of 1.5 works very well.

As a rule, I enter long position only after the price (open/high/low/close) is above SafeZone, of course after looking at other details.

With regards,

Abhay (AAD)

 

[karthikmarar]

haii karthick
iam new to tis group.i think 5 day stochastic not mush use now.all most all nifty shares stochastic are in over sold territory.can u pls tell me some good techinical indicators 4 short term treading.

i would like to start a thread which deals with the daily news analysis like fed rate etc and its impact on market. i think for intra day treading news analysis is important.
kurian

Kurian

Yes most stocks are in oversold territory as they have run up much. Now is the time for correction. May be we are already in it.

The usefulness of the stochastic depends a lot the time frame you are looking at. Just repeating from a earlier post of mine.

The stochastic oscillator is an Indicator seldom understood properly. The stochastic oscillator has really got nothing to do with buying or selling.

The stochastic oscillator tries to plot the ratio of todays level from the bottom of the selected period to the maximum level in that period. As today level rises compared to the maximum level the indicator rises. When todays level is more than the overall level of the period the oscillator peaks. As a stock is not expected to rises continuously there is a probability the stock would reverse / pull back. In others words lot of buying has gone in that stock and some profit booking could be expected. So the stock is said to be in the over bought region. Of course it is not necessary that the stock will reverse or pullback and could still rise further. So the stochastic remains in the over bought region for long till the pullback / reversal occurs.

The stochastic oscillator work fine when the market is ranging. In a up trend the stochastic oscillator gives good entry points when it moves out of the over sold region as these points refer to the resumption of trend after a pull back. So use the oversold signals for entry in a up trend. In the same way use the overbought signals for shorting in a down trend.

Also it is important at what time frame you are looking. The normal 5-day stochastic represents a very short period. This would oscillate up and down quite frequently as it represents a week on a daily chart. At the same time a 15-day stochastic on a daily chart would represent a three week range of the market. This would oscillate less frequently. So one should chose the parameters as per time frame he is looking.

About Indicators for short time frames, I am afraid I am not knowledgeable enough to comment on that as I concentrate on positional trades.

Yes, I agree that news is important for intraday trades. Unfortunately rumors too. Please do start the thread daily news. It will definitely help the Intra day traders provided the news is updated immediately.

Warm regards

Karthik

 

[karthikmarar]

Hello,


To supplement more on karthiks explanation on stoploss and moneymanagement i have attached a wordfile . It should be beneficial.

Regards

Saji

Saji

Thanks for the nice Link. Looking forward for more from you.

Warm regards

Karthik

 

[Saint]

Hi SGM,

This Thread is becoming more interesting with you, saji, marcus providing valuable information.

It very true that consistently following a simple trading system with good money management and disciple would be best way to trade.

Tough not related to pure technical Analysis, I would like to elaborate on this important topic a bit.

To give you an example why Money Management is important for trading. Let us look at how limiting your loss affect a trading system.

The expectancy of a trade is defined as

Expectancy = (Probability of win * average Money made)- (Probability of loss* Average money lost)

Suppose if we consider the probability of winning and loosing is the same , ie. 50%.

So the formula would be E= (0,5* money made) (0.5* money lost)

Assume that you will implement a strict stop loss and will restrict the loss to a maximum of 10% of your money invested. But at the same time you will let your profits run and work on a risk to reward ratio of minimum 1:2, meaning that you will make at least 20% on winning trades.

So what will be the expectancy of this system?

E= (0.5*0.2)-(0.5*0.1) = 0.1 0.05 = 0.05

What it means is that you would make 5 paise for every rupee invested. Suppose you restrict you loss to 5% then you would make 7.5 paise for every rupee invested.

Interesting isnt it? Even with the probability of win and lose being the same you would make more money if you strictly implement stoploss. This clearly brings out the importance of stoploss, which is a key factor in Money management.


It would be interesting to see if we have strict stop loss of 5% and also a system, which gives a winning probability of 60% (6:4), then you would make 10 paise for every rupee invested. At the same time if you make 30% in winning trades instead of 20%, you would make 16 paise per rupee invested. So it is important to make the maximum in winning trades (where exit philosophy comes into picture) and restrict your losses with strict stop loss. Play with different figures in the expectancy equation and you will realize how important it is to cut your losses per trade.

Warm regards

karthik

Very nice,my friend.......Great stuff,and keep em coming.

Saint

 

[rahulg77]

Dear Rahulg77,

As pointed out by Karthik in his answer to your question, trailing stop-loss is one of the ways to protect your profits (as well as capital), once prices start going up. I am giving below a formula that I use in MetaStock (as an indicator). I just move my stoploss figure to the one shown by this indicator. I call it SafeZone

I have derived it by reading "Come into my trading room by Alexander Elder", with some modifications. Expect everyone to comment on its usefulness as trailing stoploss.

-------
multiFactor:=Input("Enter factor of agressiveness",0.5,2,1.5);
stopLoss:=If(L>=Ref(L,-1),Ref(L,-1)+multiFactor*(L-Ref(L,-1)),Ref(L,-1));
Mov(stopLoss,9,E);
-------

Here, if you use higher factor of agressiveness, say 2, you will be able to lock-in max. profits but a little higher volatility during the day may square-off your long position. Generally, I have observed that a factor of 1.5 works very well.

As a rule, I enter long position only after the price (open/high/low/close) is above SafeZone, of course after looking at other details.

With regards,

Abhay (AAD)

Hi everyone,

Thank you for adding to my post. I am sorry was out and could not follow ths thread. Abhay thank you too for sharing your method. I would appreciate if u could explain your formulae so I could also undertsand what the indicator is trying to tell us. Also how u use this S/L. U use it to exit when price moves below this on intraday bases or on EOD bases. Also, if on EOD basis then u exit if closing price is below this or entire tick is below this

Rgds

Rahul

 

[aad]

Hi everyone,

Thank you for adding to my post. I am sorry was out and could not follow ths thread. Abhay thank you too for sharing your method. I would appreciate if u could explain your formulae so I could also undertsand what the indicator is trying to tell us. Also how u use this S/L. U use it to exit when price moves below this on intraday bases or on EOD bases. Also, if on EOD basis then u exit if closing price is below this or entire tick is below this

Rgds

Rahul

Dear Rahul,

This is a method to determine trailing stoploss figure once you go long, taking into consideration the volatility for each day. I use this indicator on EOD basis (but it hardly makes any difference, as explained in paras further), only for swing trading. It might trigger a sell signal on a pullback and my position is cut on that day. Let me explain line by line, starting with 2nd line.

stopLoss:=If(L>=Ref(L,-1),Ref(L,-1)+multiFactor*(L-Ref(L,-1)),Ref(L,-1));

Here, if your today's low is higher than yesterday's low then yesterday's low is added by the difference in today's low and y'day's low multiplied by "factor of aggressiveness" to find your stoploss else y'day's low is taken as your stoploss price. Once you get this S/L price, you take 9 day EMA of this S/L to smoothen it out to actually determine your stoploss indicator value for that day. It works out well as you are taking into consideration the volatility for each day (since you are taking its lowest value - the power of bears), so, if stock starts trending up, your trailing stoploss also starts going up, protecting your capital.

As I said, I use it on EOD basis. One of my buying rule is to go long only if today's price (all of open/high/low/close) is above SafeZone. So lets say, I find today that a particular stock is above SafeZone, I go long tomorrow, keeping my S/L price to be the same as today (since I cannot know what will be my S/L price tomorrow in advance, so to be on the safer side). Once I go long tomorrow, I see at the end of the day what is the value of SafeZone on my buying day and put my S/L price at this price for the next day. This way, I catch the trend if stock starts going up or my position is cut if it triggers my SafeZone price, protecting my capital. Generally, I have observed that when SafeZone is breached, the stock goes down further (this can either be a pullback or may be very high volatility beyond what we have witnessed in the past or just a reversal). Once SafeZone is breached and my position is cut, I keep a watch to see if it was really a reversal or a pullback. If it was a pullback, I re-enter again once it starts trending up again after pullback (of course if other buying signals are suggesting a buy at that time). If it was a reversal or if stock goes down and becomes rangebound, you are happy as you booked profits at higher price and that capital is free to be invested in other trending stocks.

This indicator works well for swing trading and in trending scrips. When stock becomes rangebound, this indicator is not so useful.

Last comes the first line which specifies "factor of agressiveness". I have found here that using very low values (like 0.5), healthy pullbacks - say 50% on Fibonacci Scale, do not trigger a sale signal but it also carries a risk if it was not a pullback or if it was a reversal, as you lose much of your profit by the time SafeZone is breached. Higher values (like 2) let you lock-in maximum profits but can also force you to exit early if there is too high volatility. Factor 1.5 works well for me.

I would appreciate if you can use this indicator on historical data of some stocks and see where buy signals and sell signals are generated and if they were really worth and let me know your comments.

I hope I have given in-depth working of this indicator. Please let me know if you need any further information.

As always, comments on usefulness, improving this indicator further, from everyone, are more than welcome.

With best regards,

Abhay (AAD)

 

[karthikmarar]

Hi All

Work pressure had kept me away from the thread for some time. Pressure continues and I hope to be back in full form soon.

Meanwhile there is no harm in mixing TA with some Fun. Otherwise it becomes too boring. Here is a Indicator just for Fun.

I had always been impressed with the Heat Maps I saw on the Nasdaq and some US site. Was wondering if we can create something similar for our Indicators. So I created a heat map for some Indicators for Amibroker using some codes found on the net.

I am enclosing a image of the same. It contains three ribbons one each for MACD, RSI and Stochastic K. The color and intensity changes to green when the Indicators become positive and red when they become negative.

Finally there is a ribbon which is a combination of all the three. It looks nice. It gives a nice overall story. When all the bands turn green the stock is moving up and viceversa.

I dont know if this has some real value other than the fun part. Anyway Amibroker users can have some fun with this. The afl is enclosed and the parameters of the three indicators are adjustable.

Metastock users, sorry this cannot be implemented in your favorite software. Looks like it is time you switched to Amibroker ..

regards

Karthik