Which one is the Best?

[uasish]

Though i have posted this at relevant trade ,thought of posting it here also,this would be repetation still i believe indicators are must even to garner probability edge over the coin toss effect.
1st Col= Projected Close of to-morrow based on
a derivative of today's price(named 'Mimi'
my daughter's name);
2nd Col= If Mimi inside today's Range
then 1 ( Reason is: we cant enter if it is
outside the Day's High or Low);
3rd Col = Actual close.
4th Col= Strategy:
Long if today's Close > yesterday's mimi ,
Why yesterday's Mimi becoz until today closes
cant get EOD data hence my reference is yesterday's
Mimi,which again is today's Projection only.Here plz
note though the Entry price is reflected next day ,the
Entry conditions had been met earlier & entered then.
Reflecting later to show the Profit/ Loss in Col 5 only
on Closing basis.
5th Col= Difference of Today's close with yesterday's close.
6th Col = Exit Price ,here whenever Mimi > close today ,then
Exit Price is Next day's Open.
............................................................................
This is an experiment in Metastock only no other Software
used, & the Results are derived from Metastock Exploration.
Only to see whether alone Metastock can do backtesting
without the 'System Tester' of MS.

Started on 25th June 2003
After each Entry the 1st Exit is considered here.

As done in Metastock Exploration hence
had to do the Drawdown Manually.(Could do that also if MS 10
used ,becoz it has more exploration cols)

Drawdowns:
1st June 2004 1 contract (-) 53.00
17th Aug 2004 1 contract (-) 52.50
2nd Feb 2005 1 contract (-)1.64
26th Aug 2005 1 contract (-) 43.75
19th & 20th Jan 2006 2 contracts (-)0.25
17th May 2006 1 contract (-)78.00
21st July 2006 1 contract(-)126.00
27th Nov 2006 1 contract (-)5.46
15th Feb 2007 1 contract (-)21.00
14th May 2007 1 contract (-)9.73
http://www.4shared.com/file/16693732/f710bd39/Column_Details.html

No Stop Loss used.Going thru the report plz suggest what should be my SL % on closing basis.
As Drawdowns are calculated manually may be 1 or 2 missed plz correct me.The Exit is always on 1st Exit price after the Entry price.This is a Long only ,the Exit price can be used as Short Entry & in that case 1st Entry price as cover price.

 

[beginner_av]

Seeing the illustration, will you trade this system when the blue line crosses over the red line (a short EMA crossing the long EMA)?

Well if your answer is yes, think again. This is just a cross of EMA 3 over EMA 5, as close to price itself. Do you consider a flawed system? Whether you do or dont, see the result of this trading in ABB from 2nd April till 24th May, quite a long period. This method has made quite a good amount of money. Now put this to back test over a long period with just stop and reverse and you will be ravaged. This is what I meant by even flawed system (you can call this junk) can make money and you can start off very well. Readers should understand that starting off is not the same as using it successfully.

 

[Bullet]

Moving averages smooth out a data series and make it easier to identify the direction of the trend. Because past price data is used to form moving averages, they are considered lagging, or trend following, indicators.

Moving averages will not predict a change in trend, but rather follow behind the current trend.

Therefore, they are best suited for trend identification and trend following purposes, not for prediction.

 

[uasish]

This is what can be reasonably expected with indicators ,Money mgmt & Risk mgmnt.
In Futures mkt with 2.75 Lacks (2.5 for contract size & 15k for Drawdown to start with) & on an avg of 20% margin money per trade at the most 5 trades max at a time & 2% (of 2.75L) max risked,& 200/= for each leg of contract accounted as brokerage.
Theoratically Avg 1315 trades executed since 21st May 2004.

http://www.traderji.com/92027-post101.html

This is the Equity Cuve on Random Walk Simmulation:=

http://www.traderji.com/92035-post102.html

 

[kkseal]

Would a LEADING TREND INDICATOR then be the Holy Grail?

BAV, Uasish - will be waiting for your reply.

Regards,
Kalyan.

 

[uasish]

Cant say,
kalyan last few weeks it is relentlessly being pointed out that the nature of the outcome of any trade is 'Random',theoratically ,logically & mathametically.
With my limited knowledge & experience i am truly trying to grasp the subject ,in my own way ,theoratically,logically & mathametically.
Now the ultimate was ,whether to-day's price has any thing to do with yesterday's price.
My findings,in my own crude way is YES price is Random but it has a 'Memory'
unlike coin toss,but like a game of 'Poker'.
To travarse this path all intruiging,probing posts from all of you has infact helped me to instill this in self.A protracted war with my "confussed state of mind" & self belief has now ended.Thks to all of you for helping me to stride forward in my understanding.
I am here ,in this forum to share my 2 bits,all my agonies & experience,God is kind enough & Mkt always Humbels me.
Trading has now become a wonderful ,rewarding activity truly enjoying every bit of this non lasting streak of coin toss.
Still vividly remember ,firmly etched in memory ,of those agonising ,aimless DARK years when my bank a/c slowly reduced from 6 fig to 4 fig.I had all my responsibility as Husband ,Father & Son ,& except trading there was no other source of income.Entered this Holy shrine , when i was above 45 yrs of age.
It is a long path for me,God has designed so for me,i dont need any Holy Grail,
this IS a Holy place.

 

[uasish]

This is the same like earlier but here a strict Risk Mgmnt of Fixed Dollar model of Max $ 5000.00 Risked per contract in Futures & Transaction cost of $ 200 for each leg (like earlier) but NSE margin requirement of $ 50000.00 for each trade specified in trade database code.

http://www.traderji.com/92053-post102.html


http://www.traderji.com/92053-post102.html

Here no pyramiding of Profit done & more Random simmulation (500 for each trade) done to catch the actual mkt enviourment.
Say i get 2 signals at EOD Data,i know irrespective of any thing happening (say pakistan attacks)
next day at 9.56 i am going to fire a Long order,& instantly getting the fill fire a SL order (Rs 5000.00 divided by the Lot size) minus the Entry price.Nothing more to do for entire day ,can even go to a multiplex to see a film with family member, quality time as per 'Saint's' advise,until EOD data recvd & verified the Exit signal.
In that situation of recieving 2 signals i will definetly select the one having more upside as per EW.
Will it not give me an edge ,even to beat the outcome of this machine,this role of TA can not be denied.

 

[beginner_av]

Would a LEADING TREND INDICATOR then be the Holy Grail?

BAV, Uasish - will be waiting for your reply.

Regards,
Kalyan.

No KK, it wont. Dimitris, a prolific contributor in AMibroker forum has done some work on predicting moving avg crossovers, available in TASC magazine. But that is just the price reqd for the cross to happen. Whether the price will actually cross or not, nobody knows. To answer your question, any leading indicator whether trend or momentum that predicts accurately (which is impossible till now, may be after time machine is built!), will be a holy grail. But then if that indicator predicts 100% accurately, then there wont be any trades, it will be like your FD maturity amount, you know that on X date the amount will be Y. So it wont be a holy grail any more!!!

 

[kkseal]

No KK, it wont. Dimitris, a prolific contributor in AMibroker forum has done some work on predicting moving avg crossovers, available in TASC magazine. But that is just the price reqd for the cross to happen. Whether the price will actually cross or not, nobody knows. To answer your question, any leading indicator whether trend or momentum that predicts accurately (which is impossible till now, may be after time machine is built!), will be a holy grail. But then if that indicator predicts 100% accurately, then there wont be any trades, it will be like your FD maturity amount, you know that on X date the amount will be Y. So it wont be a holy grail any more!!!

I was actually thinking in terms of a filter (not necessarily a MA) that's zero-lag (possible) and then create a leading indicator based on a derivative of that filter (the possibility of this struck me in bed last night in a semi-sleeping state .
It may not (might as well say, will not) be 100% accurate, but even a high probability would be wonderful, don't you think?

Will have to work on this. Time is a big constraint but will somehow have to find the time.

Uasish, you have been doing some wonderful work. It's people like you who act as an inspiration.

Regards,
Kalyan.

 

[joy433]

I was actually thinking in terms of a filter (not necessarily a MA) that's zero-lag (possible) and then create a leading indicator based on a derivative of that filter (the possibility of this struck me in bed last night in a semi-sleeping state .
It may not (might as well say, will not) be 100% accurate, but even a high probability would be wonderful, don't you think?

Will have to work on this. Time is a big constraint but will somehow have to find the time.

Uasish, you have been doing some wonderful work. It's people like you who act as an inspiration.

Regards,
Kalyan.

I cant understand what all of u are discussing......