Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Trading with Candlestick Daily Green and Red 15 replies

Trading by the Light of the Moon 41 replies

Making the green with price action trading 2 replies

fx|green Trading journal 27 replies

Time Zones, Day Light Savings, and Major Markets 3 replies

Trading Green Light

  • Post #1
  • Quote
  • First Post: Apr 16, 2006 5:42pm Apr 16, 2006 5:42pm
Hello Traders,
Please Evaluate this senario.



My senario.
After reading the books, surfing the forums and attending many trading webinars it all finally became clear, or at least way more clearer then before.

Then I started working on a systems that fit my trading style, and my life style. All that was learner was integrate in developing a pretty profitable system.

Finally started fine tuning and modeling up the system that I decided to trade.

It looked good, but that did not mean anything yet, so I took the next step and did a year worth of back testing.
I have developed
-100% mechanical ENTRY and Exits
-Strict and Set rules, no need for emotions.
-Risk management system that gives a lot of room to trade and free of large dawn draws
-Money management system that yields 10% per month
(it was possible to get more then 10%, however the Risk management system will not agree, so lets play it slow and steady)

After a year worth of back testing, the system came out very profitable, I did not stop there.

I started to Demo Trade the system for 2 months and , Yes it still is profitable.





One year of backtesting, 2 months of Demo Trading.

So, Traders I have a question.

Do I have the Green Light to take it live and trade the system?
It sounds like you already gave yourself a green light. How can we possibly give you an honest opinion without knowing or seeing what you are talking about?

Post what you created here and let the community give their opinion.
Hi MrMajik,
Sorry I should have been more clear.


This is a senario.
So it's a WHAT IF


WHAT IF the systems is profitable a year in backtesting, is that when one knows that its OKAY to tade it live?


SO I would like for you guys to put yourself in the place of the trader in my senario.

Whould YOU take it live?

Quoting MrMajik
It sounds like you already gave yourself a green light. How can we possibly give you an honest opinion without knowing or seeing what you are talking about?

Post what you created here and let the community give their opinion.
No, 2 months is not enough, because you might have just gone through the perfect market conditions for your system. You need to expand that demo trial period to make sure that the system trades in a wide range of market scenarios (trending, choppy, etc) and still prove profitable.

I'd say that six months of demo should make it. (avoid august though).
Quoting WTB
No, 2 months is not enough, because you might have just gone through the perfect market conditions for your system. You need to expand that demo trial period to make sure that the system trades in a wide range of market scenarios (trending, choppy, etc) and still prove profitable.

I'd say that six months of demo should make it. (avoid august though).
Yes that is understood that 2 months is not enough.

The system was infact backtested for a year, would that override the demo.



By-the-way, thank you guys for all the feed back this forum is the best
If the system is 100% mechanical then I would back test it over as much data as I could get my hands on (with a backtesting program, not manually). If it is intraday I would use back to 2001 and end of day back to 1994 (just because that is the data that I have). I probably wouldn't demo it at all - I would make sure that I could physically trade it the way that I backtested it and if I could I would get stuck in and start trading it.

Simon
Quoting DrRock
If the system is 100% mechanical then I would back test it over as much data as I could get my hands on (with a backtesting program, not manually). If it is intraday I would use back to 2001 and end of day back to 1994 (just because that is the data that I have). I probably wouldn't demo it at all - I would make sure that I could physically trade it the way that I backtested it and if I could I would get stuck in and start trading it.

Simon


What BackTesting software would you recommend?
An what Tips do you have to share with us about backtesting?



Some that are in the main stream platforms are:
MT3
MT4
VTtrader
AMIBroker
Wealth-Lab Pro
Hi,
Quoting avlasov
One year of backtesting, 2 months of Demo Trading.

So, Traders I have a question.

Do I have the Green Light to take it live and trade the system?
I would agree with Simon to backtest over as much data as you can. You can backtest 2004, determine forward testing parameters, and apply those parameters to 2005 and see if the results are profitable.

Or you can apply 2005 parameters to 2004 data.

This checks for consistency from year to year.

Best regards
Alan
How many trades did you make during the demo period? Try for at least 30 trades, or more.

Next review all those trades and see if you can explain why they won or lost. Try to figure out what changes would have been needed in the market to make a winning trade into a losing trade. If only a pip or two difference in the price would turn a winning trade into a losing trade then it's not a good trade for you, because you won't get those pips in the real market. If you eliminate the borderline winners, would it still be a winning system? What about the borderline losers?

If you eliminate the best two or three trades, what's the results of the system? Is it still profitable? What about the worse two trades?

Note that should apply the same questions to your back tested trades as well. If you don't have at least 30 trades, you need more data.
In reality, whether you are ready to take a system live has less to do with the system and more to do with you and your ability to execute the system as described.

There are many trading systems out there, that, in the long run are profitable, yet so many lose their accounts trading the same system someone else is making a tidy profit on. Do not underestimate the effect live trading can have compared to demo trading, emotion is a dangerous thing at the best of times, let alone while trading.

If you are sure you can stick to your systems rules without wavering, and you have a good grasp on what money management is and how it applies to your system, then you are ready to lose ... I mean open .. a live account.

I wish you all the luck in the world, and an equal serving of patience.
You can quit and they won't care, but you will always know.
I will say that it is not only a time issue how long have you tested, but also how many trades. The system has taken if the system only trades two times a month than two months is deftly not enough.

I have read some of your other posts, and it seemed like you are a trader. That puts a lot of thought into his decisions. So I won't bore you with little insights, as you probably know, many of them already.

If it is purely mechanical I would back tested as far as possible using some software. I usually foreword test for about three months. Most of my system take between one to six trades a week.

I then test the system for two to three months using a micro or mini account this is probably a psychological issue and helps me get confidence in what the system is telling me.
A truly wise man, always has more questions than answers.
I add my vote to those who said that you should backtest on more data than 1 year worth...1 year of data is not at all enough....You might have been testing your system over the previous years' data for the EUR/USD and it produced great results because in this particular year, the EUR/USD pair was trending downwards heavily, while if you go test it over a period of congestion, it would be non-profitable...So all you need to do is to backtest over many years backwards...

Another point to take into consideration, is your system a short, medium or long term one??? Cause if it's a short term one, you don't need too much data to backtest on, 2 years would be enough as you will be sure that almost all the circumstances will face your system....If it is a long term one, then you need more data to see how it will behave against long term trends and congestions....

All the best...


Thanks,

Nader
Quoting narafa
I add my vote to those who said that you should backtest on more data than 1 year worth...1 year of data is not at all enough....You might have been testing your system over the previous years' data for the EUR/USD and it produced great results because in this particular year, the EUR/USD pair was trending downwards heavily, while if you go test it over a period of congestion, it would be non-profitable...So all you need to do is to backtest over many years backwards...

Another point to take into consideration, is your system a short, medium or long term one??? Cause if it's a short term one, you don't need too much data to backtest on, 2 years would be enough as you will be sure that almost all the circumstances will face your system....If it is a long term one, then you need more data to see how it will behave against long term trends and congestions....

All the best...


Thanks,

Nader

The system trades on:
One Hour charts
Shoots out about 20/25 signals per month per pair
Hold time is 10-20 hours If locked on a trend, it will follow.


I think by trading pairs that are not directory inter dependant is a benefit for choppy markets.
  • Post #14
  • Quote
  • Last Post: Apr 17, 2006 10:07pm Apr 17, 2006 10:07pm
So, What have you decided to do? Are you going to back test some more? Are you going live with a mini acct?

Thanks SMJ


Quoting avlasov
The system trades on:
One Hour charts
Shoots out about 20/25 signals per month per pair
Hold time is 10-20 hours If locked on a trend, it will follow.


I think by trading pairs that are not directory inter dependant is a benefit for choppy markets.
0 traders viewing now