[Archive] Learn how to make money villagers! - page 237

New comment
 
paukas:

I told you, you will find what you are looking for.

You will get there.


I just remembered something else = this tester stuff works on any currency pair and on any timeframe....

I have it on eight currency pairs - so far so good, what's next God only knows ....

but i am just saying....

 
khorosh:

My newly minted expert. I made it after analysing Ilan's shortcomings. The drawdown is still a bit high, but I am still working to reduce it and there are prospects for this. Not for sale.

I have tested it without reinvestment. Minimal order lot size is 0.02, maximal one is 0.2. Arithmetic increase of lot size is used.

Strategy Tester Report
e-Bomber
Alpari-Demo (Build 409)

SymbolEURUSD (Euro vs US Dollar)
Period1 Minute (M1) 2009.06.01 00:00 - 2011.12.02 23:59 (2009.06.01 - 2012.05.01)
ModelBy open prices (only for Expert Advisors with explicit bar opening control)



Bars in history927076Modelled ticks1850513Modeling qualityn/a
Chart mismatch errors0




Initial deposit1000.00



Net profit23758.29Total profit31592.18Total loss-7833.88
Profitability4.03Expectation of winning6.11

Absolute drawdown357.91Maximum drawdown3269.01 (16.56%)Relative drawdown57.92% (883.64)

Total trades3886Short positions (% win)1920 (84.17%)Long positions (% win)1966 (85.50%)

Profitable trades (% of all)3297 (84.84%)Loss trades (% of all)589 (15.16%)
Largestprofitable trade288.76losing deal-92.75
Averageprofitable deal9.58Deal loss-13.30
Maximum numbercontinuous wins (profit)34 (136.50)Continuous losses (loss)11 (-228.79)
MaximumContinuous Profit (number of wins)1115.75 (11)Continuous loss (number of losses)-309.24 (7)
Averagecontinuous winnings8Continuous loss1




Yuri, you are, as usual, that Avalanche branch, "ahead of the planet"... :-) Have to catch up with you all the time... :-) - See end of this thread page. Good for you. Can you share with us the description of this TS?
 
elmucon:


...

Don't be in a hurry to charge real, let it sit on the demo for a while longer (just a good tip, you can ignore it) .

...I'm already ignored.
 
test with a fixed lot. If there is at least a minimum sustained profit there, then martin, averaging or other types of MM makes sense to apply to enhance system efficiency. If there is no stable statistical advantage with a fixed lot no MM will bring it - this is the basics :)
 
Roman.:
already in ignore.

Well, that's the answer I was expecting .... I'll see you in Hawaii then, or do you prefer other islands...?
 
elmucon:

That's what I expected .... See you in Hawaii then, or do you prefer other islands...?


This time I prefer other islands, this time I prefer Guatemala... :-)

"This year we only had one camp in the Dominican Republic, in a very good hotel, which I love. But next year, in January, we will have a camp in Guatemala: there are no places, everything is sold out. In May 2012 we will have a Pacific camp, which usually happens every two years.

"no room, all sold out" - whatever, we'll be there, we'll have a look, I don't see the point in storming it yet... If only there was something to "break a spear" for... :-)

 
Avals:
... - it's the basics :)

I thought so myself... :-) And when you develop a trend TS based on MM on a Martin, you get a picture that you do not need such a MM - it is perfectly good without it, especially when a more or less competent TA or other - what kind of market analysis, you have to wait for the conditions for the placement of one or another order in the market and start with a minimum amount of martin, but it is not so ... When you write about MM martin with one or another kind of market martin, what type of TS you mean - the first or the second?

Reshetov writes on 265 pages of the same branch: "What is the point of adding martin to profitable TS? To speed up depo draining?

If we want to profit profit trade, there is a more efficient MM with increasing of a position after profit and decreasing of a position after loss. As a last resort, if expected payoff is small, you may earn with a fixed lot.

Look at my links to the description and video of the TS on the previous page of this thread.

 
Roman.:

I thought so myself... :-) And when you develop a trend TS based on MM on a Martin, you get a picture that you do not need such a MM - it is perfectly good without it, especially that a more or less competent TA or other - what kind of market analysis, you have to wait for the conditions for the placement of one or another order in the market and start with a minimum amount of martin, but it is not so ... This is what you see on this page. When you're writing about MM martin with one or another type of TA in the market, what type of TS you mean - the first or the second?


the second one is closer. The first is a special case because the signal to enter a new position does not necessarily coincide with the signal to exit the opposite position. It may coincide for reversal systems.

In general, martin as an increase in the lot in a trade only on the basis of the fact that the previous trade was a loss-making or profitable - this is the way to nowhere. In the best case it is a mere fitting of the series, and in the worst case it is waiting for the loss till the margin call. Increasing the lot size only makes sense if the new signal is "better" than the previous one in terms of profit/risk. That is, not all signals are equal - there are stronger (and usually more rare), which makes sense to trade with a larger share of the deposit. This can lead to averaging or pyramiding or even a soft variant of martin. This is a consequence of the fact that the system takes into account different trade potentials. But all the same, such systems should have a steady profit even when trading with a fixed lot.

 
Avals:


... But all the same, such systems should have stable profits even when trading with a fixed lot.


I was thinking and counting the same way until recently... :-) With classic martin - it does, but there are also some NUTS! !!

What conclusion I came to - in the penultimate post 263 page of this thread, on Monday I will lock the TS-ku on this video - the link in my (penultimate) post of this page of this thread ... In the tester, all draws beautifully from the beginning of quotes history from 2001. - Before 12.2010 - optimization of one APR period parameter for din of channel width to kr line, before 12.2011 - forward. Lot - 0.1. Other variables are not used in this TS in any way. No trawl. Number of turns 15 is also by default.

Channel width in pips on a five-digit with ATP period = 190 on days is obtained on the whole testing period, in general, from 400 to 600 pips.

Entry conditions are no longer on the sensor random numbers, but on the colour of the candle, because Avalanche- TREND! :-)

//открываемся стартовым лотом в зависимости от цвета свечи 
    if (iOpen(Symbol(),signal_period,1)<iClose(Symbol(),signal_period,1)) WmOrderSend(Symbol(), OP_BUY, Lots_New, Ask, 0, 0, "старт", MagicNumber);
          else WmOrderSend(Symbol(), OP_SELL, Lots_New, Bid, 0, 0, "старт", MagicNumber);
 

But my owls from 2000 I don't play with parameters there is no point) and anyway it is indicatorless... Not for SALE! DDD if you do not increase the lot at no risk but a little to someone who is not enough to the prosecutor will add :D

1...230231232233234235236237238239240241242243244...882
New comment
Reason: