Settings Files - Page 4

fikko · #31 (**permalink**) 12-15-2006, 05:16 PM

Quote:

Originally Posted by vincethebeast

Yep, I know the feeling...

Been rather busy (behind the scene this time) the last few weeks doing similar things. Still running 9 different forward tests as we speak but I find it hard to say which strategy works best..
I have done half yearly backtest, 2 monthly backtests etc etc. All with different results. The problem remains that it is hard to tell what works best in the long run. Results of just a few weeks are not telling us a lot I think... On top of that we have the difficulty in choosing a version and stick with this one for a while to see difference in settings (and not just differences caused by version changes...)

Will keep everyone posted,

Vince

I also have completed forward testing with 2 weeks historical timeframe and also 1 month, both are not good. 2 months time frame is better. 3 months setting will have similar values with 2 months.

I still have to do more forward testing scenarios. No conclusion at this time.

daraknor · #32 (**permalink**) 12-15-2006, 05:56 PM

I agree we need to stick with one version to test and optimize. I want to create a stable version so we aren't optimizing bugs anymore. :/ Phoenix 5.7.0 may be the last version for a while if it is stable... I just want to users not get Order errors every week

I replaced over 60 lines of sometimes working code with 3 lines of always working code

We'll see how it holds up in forward test.

vincethebeast · #33 (**permalink**) 12-15-2006, 09:46 PM

Great, got one ready to go for next week.

Stability is one of the reasons why I'm still running various 4.2.03 versions :-) Call me old fashioned but it works, so why not :-)
Hopefully your 5.7 will be just as stable since I always have been a big fan of mode 3 and I do miss it in Phoenix 4!
Can't wait to kill my pc's little brain with optimization runs again :-)

CU Vince

Quote:

Originally Posted by daraknor

I agree we need to stick with one version to test and optimize. I want to create a stable version so we aren't optimizing bugs anymore. :/ Phoenix 5.7.0 may be the last version for a while if it is stable... I just want to users not get Order errors every week

I replaced over 60 lines of sometimes working code with 3 lines of always working code

We'll see how it holds up in forward test.

cockeyedcowboy · #34 (**permalink**) 12-15-2006, 11:31 PM

Quote:

Originally Posted by daraknor

Those tests with random number generator proved you are full of crap. Say something concrete or stop posting in *every* thread saying the same message of vague doom. No one listens to "Chicken Little."

I never said any thing that would be considered vague doom, but on the contrary, the system can be made stable and very profitable. I have a system that is very similar and it works very good. There are a few things that the Phoenix does differently, but generally there function the same. I would not be following this thread if I though it was a wate of time. I really think that if it wont for you this thing would progress. Lets see how long it going to take you to catch on.

If you have done the tests (correctly) I assure you, you would be looking at how to elimiate the need to optimizing every thing right now. But you want every thing handed to you, what a shame. Why do I keep posting its may be some one else will start looking for the answers.

"Chicken Little", you need to grow up.

The CockeyedCowboy

paulrus · #35 (**permalink**) 12-17-2006, 01:06 AM

i am trying to understand the logic of this program as i run the strategy tester or a demo account. what would help me would be: does someone know what indicators one would set to create a template based on the settings in the program? often a trade is executed and i don't really understand why.
any guidance would be appreciated
thanks
paul

daraknor · #36 (**permalink**) 12-17-2006, 09:24 AM

Quote:

Originally Posted by paulrus

i am trying to understand the logic of this program as i run the strategy tester or a demo account. what would help me would be: does someone know what indicators one would set to create a template based on the settings in the program? often a trade is executed and i don't really understand why.
any guidance would be appreciated
thanks
paul

Paulrus, there are two recommendations for you to work from. The indicators are very clearly separated in code but they are used in unconventional ways. Attempting to plot them on a chart can work, but you won't see the signals.

The second recommendation is to use "Phoenix5_plotting_buy_sell_signals.mq4 " from the FAQ thread.

The third recommendation is to use Phoenix Mode1, set MaxTrades = 10 or 20 on a 100k demo account, turn off money management and look at it visually.

To try to save you a little time, Phoenix is a trend reversal strategy based on oscillations in average price, change in price, and other factors.

In case you were wondering about random signals, my research and mathematics dictate that a purely random entry would loose money at a rate equal to the brokers spread divided by the weighted sum of TP and SL as a percentage, calculated per trade over the long term. In simpler math terms, purely random (and pure MACD) loose money on even TP:SL bets (because random = bet) at a rate equal to: (spread in pips/ take profit)/100 * trades. You can easily test this with the MACD sample, and turn off trailing stops. Anyone claiming long term profits on random signals is lying.

If you want to completely separate the indicators from the EA, it is important to realize that one of them is purely custom and was previously highly praised by people who now claim the opposite in this thread. The other indicators are used in in ways that makes graphing them nearly impossible.

I recommend optimizing signals with Mode1 with higher maxtrades even if you trade with 1 maxtrade, or use Mode3.

daraknor · #37 (**permalink**) 12-17-2006, 09:26 AM

If Phoenix 5.7.0 is stable, we will be turning a lot of attention toward optimization and verifying settings files. It also helps to optimize on the data from broker you trade with.

cockeyedcowboy · #38 (**permalink**) 12-17-2006, 03:25 PM

Quote:

Originally Posted by daraknor

Paulrus, there are two recommendations for you to work from. The indicators are very clearly separated in code but they are used in unconventional ways. Attempting to plot them on a chart can work, but you won't see the signals.

The second recommendation is to use "Phoenix5_plotting_buy_sell_signals.mq4 " from the FAQ thread.

The third recommendation is to use Phoenix Mode1, set MaxTrades = 10 or 20 on a 100k demo account, turn off money management and look at it visually.

To try to save you a little time, Phoenix is a trend reversal strategy based on oscillations in average price, change in price, and other factors.

In case you were wondering about random signals, my research and mathematics dictate that a purely random entry would loose money at a rate equal to the brokers spread divided by the weighted sum of TP and SL as a percentage, calculated per trade over the long term. In simpler math terms, purely random (and pure MACD) loose money on even TP:SL bets (because random = bet) at a rate equal to: (spread in pips/ take profit)/100 * trades. You can easily test this with the MACD sample, and turn off trailing stops. Anyone claiming long term profits on random signals is lying.

If you want to completely separate the indicators from the EA, it is important to realize that one of them is purely custom and was previously highly praised by people who now claim the opposite in this thread. The other indicators are used in in ways that makes graphing them nearly impossible.

I recommend optimizing signals with Mode1 with higher maxtrades even if you trade with 1 maxtrade, or use Mode3.

I have a friend that is a commercial real estate appraiser. He does work on some of the largest real estate developments arround. One day we were having a few drinks and he told me how he can demand such high fees for his work. To be good in this business he says, one has to "impress the client with your brinllance" and he said that at times when he hasent a clue to what to do, he falls back to his only rule in appraisal, "If you haven got a clue, you just baffle them with a whole lot of B.S." that right you just say a whole lot of things that sound to complex to comprehend by the average person and people will beleive you, cause how can anyone say all that without knowing what he is doing.

I have seen you do that on a number of occations when your stuck. LOL.

The Cockeyed Cowboy

cockeyedcowboy · #39 (**permalink**) 12-17-2006, 03:41 PM

Quote:

Originally Posted by daraknor

....In case you were wondering about random signals, my research and mathematics dictate that a purely random entry would loose money at a rate equal to the brokers spread divided by the weighted sum of TP and SL as a percentage, calculated per trade over the long term. In simpler math terms, purely random (and pure MACD) loose money on even TP:SL bets (because random = bet) at a rate equal to: (spread in pips/ take profit)/100 * trades. You can easily test this with the MACD sample, and turn off trailing stops. Anyone claiming long term profits on random signals is lying.....

A little bit of knowledge can be a handy cap at times...... This is why all the fancy math and the big degrees dont make for a good trader.

spread in pips/ take profit)/100 * trades I beleive this is what Bush was refering to when he talked about fussy math. Some of what you said above is true, if your useing a randam data series, like trowing a set of fair Dice in a fixed odds game. How ever the market is far from random at times.

You may not remember or not in the business back in the 90' when the Wall Street Jruonal on the first of every year was haveing monkeys pick stocks by throughing darts at a stock list. Those monkeys were doing better then 99% of the best stock pickers arround at the time. Now that is about as random entries as you can get. It has been proven that a random entry can be profitable with good exits. That is a fact my friend. Just like a good entry and a random exit can be profitable, at times. There is one more situation you have to consider. Random entry and random exits at times can also be profitable, how? Look at the WSJ's monkeys again they picked stocks at random, the exit was the first of the following year, again, a random exit. So how did they make money? That is were the data set (maket condition) comes into play. You see that when the market is trending up and 90% of the equities are also trending, all one have to do is go long any stock at random and chances are your have 90% winners no matter how your exits are. That profit should be equal to the increase in the market (say the Dow) minus the cost of trading (spread, commissions, etc), in ' general '.

What does that have to do with the Pheonix? If you would have done the tests as stated, you would of found out which parameters that needed to be optimized. and which ones have very little to do with the outcome. Nothing more or nothing less. It had nothing to do with whather the signals were good, bad or indifferance. Just to show which varibles to optimize.

If you would stop and try to find out why the pheonix works the way it does, you would have the knowledge as well to make those parameters self adjusting. I will give you a clue, ask you self when does the Pheonix profit the most? When does it profit the least (losses)? Now ask yourself which parameters control those profit swings and why? You will see it has nothing to do with your fussy math but with the data set itself. If you can figure this out you will be able to see the hart and soul of the Pheonix.

Why does the Pheonix work better with some currency pairs and not so good with others?

Why does the Pheonix work better with different brokers? Are those brokers crooks?

This is like beating you to the side of the head with a brick. Step back man and look at it again as a third person. Learning must come from within yourself, you will never learn by being told, WHAT IS. If you can see WHY the WHAT, HOW, WHEN, will all come easy. Being told what to expect, will always leave you at the mercy of others. This is one of the problems with schooling, it leaves the students with the inpression that every thing in life will be told (given) them. Drop that 100 function calculator of yours and use common sence.

The Cockeyed Cowboy

SOMETIMES, A LITTLE BIT OF KNOWLADGE IS A GOOD THING TO WASTE.

cockeyedcowboy · #40 (**permalink**) 12-17-2006, 03:47 PM

Quote:

Originally Posted by daraknor

......If you want to completely separate the indicators from the EA, it is important to realize that one of them is purely custom and was previously highly praised by people who now claim the opposite in this thread. The other indicators are used in in ways that makes graphing them nearly impossible......

Below is, in part, the very first post I made to this thread. Which I later removed. It goes on to say what I have been saying all a long. I have not changed my views. It is you that only hears what you want to hear. If any one wants to see the complete post I will repost it.

Quote:

Originally Posted by Cockeyed Cowboy

.....The 5 Singals are, (not in the coded order);

1..There is a trading time filter that one can set to start trading and stop trading at any time that he wishes....the contest code is set to trade 24 hours continually.

2..There is a trend filter which denotes the current trend in the market. It will look to position the system againt that trend.

3..Theres an indicator that shows the current price extention of the trend (overbought oversold condition), that is, is the trend over extended in its current trend direction. If so that its again a signal to establish a position against the trend. (this over extention must pass a threshold level)

4..There is another indicator setup that is designed to indicate treanding conditions. In which Hendricks reversed the parameter list to give signal against the under lying trend.

5..The last setup is a very clever peice of code that measures if there is a pause in the trend. Kind of like looking to see if the current trend has blinked, indicating that they may be truble in the current direction. This again is a signal to enter market against that trend. There is more to this code then meets the eye!

He uses a position sizing (MM) that has a user varible input to adjust the amount of risk. It also has a feather that decrease the risk as the system experences losses. This is a very un-common design feather to most available EAs.

That is the system in a nutshell, It is a trend reversal system. The use of the different setups and filters are good and the coding style is also better then most I've seen. However there are some underlying problems with the current system. Its like playing russiam roulette with 2 bullets in your gun. These situations may be the reason that some people are acheiveing vering results. This by no means should be viewed as negitive or a put down to Hendricks. He has put in a lot of time in this. My only reason for writing the remainder of this post is to give my views on improvements to the system.

.......

The Cockeyed Cowboy

[edit] This was posted when the Pheonix was in number one positions with just under 30,000 in account. Way before it started to stumble. currently at around 22,000.

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Change MT Default Settings	zenfx	Metatrader 4	2	06-12-2007 01:32 PM
Alter time settings	droesparky	Indicators - Metatrader 4	1	06-05-2007 11:12 PM
EA settings	european	Expert Advisors - Metatrader 4	2	03-08-2007 07:09 AM
Hide Settings	pabloski	Metatrader 4	1	03-01-2007 09:26 PM
What is the best settings on EuroFx2 EA?	cnsnmrt	Expert Advisors - Metatrader 4	1	02-23-2007 10:57 AM

Register in Forex TSD!
Trading Systems Leaders in this forum (automated trading systems) are winning more than 3000 pips in a month (30000$ investing one lot every time). Click here to register and get more information