Ward 6 - page 4

[Pavel Pronin]

TO INCREASE THE PROBABILITY OF AN UMP IS NECESSARY:

1. USE AT LEAST THE SIMPLEST INDICATOR, FOR EXAMPLE, THREE-PERIOD STOCHASTIC - IN NORMAL HANDS IT WORKS STEADILY

2. USE AN NDICATOR 5 TIMES BIGGER THAN A SL - THE PROBABILITY OF AN NDICATOR IS HIGHER

[[Deleted]]

Dr.Drain:
No, no. To understand that your filter is not ahead of anything, consider it on simple signals instead of price: meander, sine wave, triangular sawtooth, and the main thing: a step. That is where you will see the lag. For miracles do not happen, and you cannot predict in advance where a pullback will occur and that there will not be a pullback.

That is not what I meant.

You can't get ahead of what doesn't exist - everyone knows that.

You can only assume and hope.

[[Deleted]]

Savl:

Is item 2 together with item 1 or can it be separately?

[[Deleted]]

DmitriyN:
Is item 2 together with item 1 or can it be separately?
:)))

It's just that the person hasn't clarified what she's all about "TP probability" - that's where it all falls into place.))))))

[[Deleted]]

Roman100:

Roman, you put TP=SL in your trending system. What will happen to the MO in this case?

[[Deleted]]

DmitriyN:
Roman, you put TP=SL in your trending system. What will happen to the MO?

To be honest, I still don't understand how the MO is calculated in the tester. Do you know?

Without the SL and TP calculation block...it might go down though...

here... MO=47

[Aleksander]

But if - for the same year 2008 - you change tp<sl slightly - and condition - Selkie at a price lower than the long MA... and vice versa...

then the result would improve...

===

Symbol		EURUSD (Euro vs US Dollar)
Period		5 Minutes (M5) 2008.01.02 10:00 - 2008.12.31 17:55 (2008.01.01 - 2009.01.01)
Model		All ticks (most accurate method based on all smallest available timeframes)
Parameters		Lots=0.1; BaTP=400; BaSL=500; SeTP=400; SeSL=500;
Bars in history	74639	Modelled ticks	4156824	Simulation quality	90.00%
Chart mismatch errors	0
Initial deposit	10000000.00
Net profit	13074.32	Total profit	149450.40	Total loss	-136376.08
Profitability	1.10	Expectation of winning	6.86
Total trades	1905	Short positions (% win)	960 (53.96%)	Long positions (% win)	945 (53.23%)
		Profitable trades (% of all)	1021 (53.60%)	Loss trades (% of all)	884 (46.40%)

[Vladimir Paukas]

Roman100:

To be honest, I still don't understand how MO is counted in the tester. Do you know?

Without the SL and TP calculation block...it might be a waste, though...

here... MO= 47

Mo in the tester counts as everywhere else. The final profit is divided by the number of trades.

[Denis Timoshin]

Roman100:

To be honest, I still don't understand how MO is counted in the tester. Do you know?

Without the SL and TP calculation unit...it might be a waste, though...

here... MO = 47

As practice shows, this number of trades shows nothing

[Rorschach]

Trading noise from the average). There was such a thought, came to a standstill. Max what I managed to get is the optimal smoothing/lag ratio, gives the JJMA indicator. Put it up a notch and use the indicator from the digital method generator to try and find something better.

There may be a loophole in the positive GVZ, haven't dug in that direction. Divergence on oscillators as an example of this effect.