Experiments in Technical Analysis

[karthikmarar]

Dear Sanjay

Excellant suggestion. This will give a real feel of implementing a system and members can learn a lot from it. I am fully on this with you..

Stock filtering

Volume could be added to the criteria. Ofcourse the FNO stocks would meet this criteria. If we want to included other stocks we could consider volume. May times we get stuck with illiquid stocks. I have had this experience with Avaya Global..

After hearing more views we can set the stock filtering criteris.

Use of leverage

FNO, why not... we should definitely consider

To start using the system we should have a owner for the portfolio who will run the show. Obviously the choice is none other than yourself.

I suggest it is better not to go full throttle. As with any new system better to start in the first gear and change gears as we understand the road ahead..

Again.. which signal to choose? Here again we better set some rules... no place for emotions. Let me put my thought on it while others also share their views.

Thanks for the great idea .. we will all learn a lot from this..

warm regards

Karthik

 

[murthymsr]

dear sanjay, karthik and others,

buy/sell signals are one part, but the more important aspect is the set of interested stocks one may like to trade at all. this is entirely one's own perception of risk taking readiness and returns expectation.

all the stocks traded in the market are to be subjected to this criterion and the buy/sell signals may be searched within that set only.

i would like to apply filters on the price, volume and the turnover to arrive at the resultant set of tradeable stocks. this could be a dynamic list unless the horizon is long term. a typical set consists of close>20 AND volume>10000 AND close*volume>10000000 (Rs 100 Lakhs).

an abstract from my trading set filter is presented below. this will allow the user to set his own criteria and arrive at the tradeable set of stocks.

//Parameters
MyC = Param( "CLOSE", 20, 0, 100, 10 );
MyV = Param( "VOL", 10000, 0, 100000, 10000 );
MyL = Param( "Rs LAKHS", 100, 0, 1000, 100 );

// Inclusions
include = (EMA(C,5)> MyC AND EMA(V,5)> MyV AND EMA(C,5)*EMA(V,5)/100000> MyL ) AND IndustryID()>4 ;
...........................
filter= (BUY OR SELL) AND include
...........................

the industryID filter in the above quoted text will allow only allow the 'stocks' to be included and NOT the composites. the industyID's are mostly based on the thread on StockIndustry assignments in AMIBROKER:

http://www.traderji.com/amibroker/8686-amibroker-database-nse-stocks.html

karthik's MABIUTS or any other buy/sell criteria logic may be searched within the above filtered set and resultant list may be seen in the exploration mode. whatever the buy/sell logic you may adopt, it may be preferable to have a look on 'other' parameters like RSI, volume surge, support/resistance and stochastics.

how much to buy may be governed by the money management principles when the buy signal is generated and when to SELL by the MM criteria & the SELL logic. it may be prudent to keep a low profit target and low stop-loss limits even that may mean 'under-utilization' of the strength of the move.

all the best.
murthymsr

 

[SGM]

buy/sell signals are one part, but the more important aspect is the set of interested stocks one may like to trade at all. this is entirely one's own perception of risk taking readiness and returns expectation.

all the stocks traded in the market are to be subjected to this criterion and the buy/sell signals may be searched within that set only.

i would like to apply filters on the price, volume and the turnover to arrive at the resultant set of tradeable stocks. this could be a dynamic list unless the horizon is long term. a typical set consists of close>20 AND volume>10000 AND close*volume>10000000 (Rs 100 Lakhs).

an abstract from my trading set filter is presented below. this will allow the user to set his own criteria and arrive at the tradable set of stocks.

Hello

Thanks Murthygaru for your inputs.

My suggestion of using only selected FnO stocks for the exercise was precisely for those very reasons. Besides liquidity and lower slippage, we would need a smaller set of stocks to choose from as our money management rules constrain us to go for maximum 5 simultaneous trades (with max 20% per Script).

karthik's MABIUTS or any other buy/sell criteria logic may be searched within the above filtered set and resultant list may be seen in the exploration mode. whatever the buy/sell logic you may adopt, it may be preferable to have a look on 'other' parameters like RSI, volume surge, support/resistance and stochastics.

We can have many rules, but we should make them objective and we should try to keep out discretional choice while selecting buy/sell signals. If you can suggest the criteria for the 'other' parameters you mentioned RSI, volume surge, support/resistance and stochastic, we will incorporate them while selecting the trades from amongst the signals generated by the original system.

how much to buy may be governed by the money management principles when the buy signal is generated and when to SELL by the MM criteria & the SELL logic. it may be prudent to keep a low profit target and low stop-loss limits even that may mean 'under-utilization' of the strength of the move.

We will have basic rules for money and risk management in place but, they can evolve as we go along.

I will try to write down a detailed plan, which we will update using the inputs from other members.

Warm Regards
Sanjay

 

[murthymsr]

dear sanjay,

We can have many rules, but we should make them objective and we should try to keep out discretional choice while selecting buy/sell signals. If you can suggest the criteria for the 'other' parameters you mentioned RSI, volume surge, support/resistance and stochastic, we will incorporate them while selecting the trades from amongst the signals generated by the original system.

what i intended to say was that the other parameters also may be looked into for a confirmation of the signal and did not say to include the other parameters into the buy/sell signals.

we may have a look into the 'other' parameters through the exploration mode, with addcolumn command, like:

AddColumn(RSI(14),"RSI",3.0,textcolor=IIf(RSI(14)>70.5, colorRed, IIf(RSI(14)<30.5, colorBlack, colorGreen)));

one may add many such columns in the exploration mode to have a confirmation of the signal. inclusion of these in the logic may not be worthwhile.

hope i am clear at least now.

murthymsr

 

[asnavale]

Hi Karthik,

Thanks for the info. With all NSE stocks (more than 1000) over a period starting from Jan 05 to Nov 06, the number of trades is not at all high. But it would be only an academic excercise to trade the whole NSE list. If we can limit the number to a reasonable level it would be more useful. Also, by eliminating insignificant stocks we may be able to increase the success ratio considerably. This does not mean I am down grading your excercise. Sometimes, such tests - though they do not reflect the real situation - are required because they may spring some surprises. In this view your test was worth the time spent. A little more tinkering may give us some ideas.

The idea proposed by Sanjay and supported by you and Murthy is a very good suggestion. I would be eager to participate in it. I also support the view that we should progress step by step.

However, I have one drawback. I do not have any experience in FnO. I will therefore be able contribute in cash segment trades. I am trying to learn the derivatives trading but not very confident. I may be able to learn easily and rapidly by following this combined excercise.

Looking forward to more developments in this proposal and waiting eagerly to partcipate in it.

Regards

-Anant

 

[murthymsr]

dear friends,

further to my earlier post on filters to arrive at the set of tradeable stocks, i am attaching the AFL and the selected list for the EOD data upto 2006/11/23.

the AFL is presented only to make the concept clear and may work on your system ONLY after some changes.

regards.
murthymsr

 

[SGM]

dear friends,

further to my earlier post on filters to arrive at the set of tradeable stocks, i am attaching the AFL and the selected list for the EOD data upto 2006/11/23.

the AFL is presented only to make the concept clear and may work on your system ONLY after some changes.

regards.
murthymsr

Dear Murthygaru

I will go thru the files you have uploaded. Hopefully we can complete the planning during the weekend.

Thanks God for Sundays (and yes Jummas also)

Warm Regards
Sanjay

 

[SGM]

"If you fail to plan, then you plan to fail". I don't know who first said that, but it's a very sound piece of advise indeed. Planning is something that is all too often overlooked by traders, and yet a well drafted trading plan is one of the most important tools for success and profit.

In talking to struggling traders, I am constantly amazed at not only how many don't have a trading plan, but how many don't even know what such a plan is. In fact a trading plan is quite simple, it's a document that details every aspect of your trading strategy. It is literally a blue-print for your trading methodology.

What should be in this document? Here are the most important areas it should cover:

• Mission Statement - A defined objective for your trading; if you don't know what it is you are trying to achieve, how will you know when you have achieved it? Having a well defined goal is essential to success in any venture.

• Pre-market preparation. Actions required before the market opens, setting up for the trading session, reviewing economic calendars, and so on.

• Trade entry rules

• Trade exit rules

• Money management rules

• Post market actions - trade logging and analysis.

As anyone who has traded in a live market knows, we must often overcome our natural emotional responses in order to execute our trades correctly. Cutting losses and letting winners ride can be easier said than done. By defining as precisely as possible, our criteria for entering and exiting trades, we have a reference that we can use to help us overcome these responses.

The trading plan should be kept at hand throughout the market session. When we see a possible entry coming up, referring back to our strict written entry criteria we can objectively look at the chart and make a informed decision about whether to enter or pass. The same applies to exiting, whether the trade is winning or losing. Over time, trading becomes almost mechanical and stress-free.

Action: We must have a written plan that defines all aspects of our trading, and we must commit to following it to the letter. Only by rigidly sticking to our strategy can we honestly determine if any problems in our trading lie within the system itself or within our execution.

Dear friends

Plz chk out the original post, one of the many gems on traderji.

Mistake Number Two - Not Having a Trading Plan

The other parts of the series

Mistake Number One - Switching Strategies or "The Hunt For The Holy Grail"
Mistake Number Three - Not Understanding Money Management
Mistake Number Four - Not Testing
Mistake Number Five - Not Putting In The Required Effort
Mistake Number Six - Overcomplicating It
Mistake Number Seven - Not Taking Action

Warm Regards
Sanjay

 

[karthikmarar]

Hi! Karthik,

Could you please share with us what is the system that you use for Buying & Selling stocks.

1) What are the parameters that you use to shortlist stocks
2) What is it that tells you that this is the stock that I am buying
3) Do you use any indicators to shortlist stocks or to check overbought / oversold levels

If you could please let us know in detail, it would be helpful to us.


Thanks & regards,
Amit

Hi Amit

I just look at price and volume.

Just Kidding This seems to be the fashion to say this nowadays.. Many have said this in this forum to newbie queries. Of course none cared to explain to the newbie what they would look at in detail and the role-played by volume.

I do use some indicators. Normally I run a few scan to short list the stocks. Final decision is taken only on visual inspection of the chart. The scans include some of my custom indicators in addition to some based on Trix, Linear Regression and even Mabiuts.. These scan and stock selection are tuned to my trading style. Trading for me is part time and it would not be incorrect if I call it a hobby. A hobby, which could pay for your wifes designer kitchen, is not bad at all. As I cannot be in the market all the time I look for stress free trading with picks which have more chances of winning and hence the selection is more conservative. The strategies are basically trend following. Right now it is the preparatory period for my eventual return to my beloved home country when I plan to start full time Trading.

These custom Indicators are still under study even though they are giving good results for me. Some are these were passed on to some members for beta testing. Since I have not heard anything from them I presume that they have found these useless and feel bad to tell me so.

I know I have not given the Answers you are looking for. I do wish I could put down everything in one post. I do plan to share these as and when I am fully satisfied with their performance. Soon I will definitely start sharing some of the strategies based on common indicators and then follow up with the custom ones. Please bear with me till then.

Unlike the views mentioned by one of our friends if I find a good trading system I will not hesitate to share it with the forum. After all I am not competing with the few members in the forum. We are all small retail traders who try to take the bits from the sharks. In fact a system shared would be more effective as it will become self-fulfilling, as many would start trading with it. Sorry for all the rambling.

Warm regards

Karthik

 

[SGM]

Hello friends

The Trading Plan

Objective: (Mission Statement)
To try to earn a profit of n% per month consistently on the capital deployed while keeping the drawdown periods to minimum. (plz propose a value for n)

Money Management Rules
Maximum of 20% of available capital should be used as margin for any one trade.
Maximum 5 trades should be open at any given time.
Maximum loss per trade should be kept below 3.33 % of available capital.
Initially a money stop loss of 5% of the position size will be used.
On the position moving in favor by more than 5% a trailing Stop of 10% will be used.
Stoploss will be modified everyday at opening.
No predefined profit targets have been set, trades will be squared only on exit or stoploss signal.

The above rules will limit the maximum leverage to 3 times of the capital available.

Trade Entry/Exit rules
The current candidates are
1. MABITUS H
2. MABITUS K
3. MABITUS K1
4. MACD Dual Timeframe (Weekly/Daily, also proposed by Karthik)
5. Plz propose your systems here.......

The entry/exit signals will be available at EoD after the data is downloaded and the scan/explorations run (will use the one uploaded by Murthy, few posts previous to this).

Trade Execution will be done the next day at opening.

Transaction cost will be considered to be 1% for cash and 0.15% for FnO for both sides of the trade.

Pre market preparation:
(Actions required before the market opens, setting up for the trading session, reviewing economic calendars, and so on.)
Inputs required. . .

Post market actions
(Trade logging and analysis.)
Inputs required. . .

Currently all the systems proposed are based on Weekly and/or Daily timeframe. If any member would like to propose something that considers intraday data (JDM’s Idea), it would be a great learning experience for all.

I hope this will qualify as a initial draft for the trading plan. Members inputs awaited to convert it into final draft.

I would request Anant to take up the implementation of MABITUS H.

If more members are willing to volunteer we can take up many different implementations.

Warm Regards
Sanjay