Experiment - page 70
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Right, it may be 1 in 40, then the number of profitable trades is minimal, the main thing is that the global average trend is able to overlap the current losses, otherwise it will just be meaningless...
You bet you can! But maybe not in all DCs.
There's a red slant there. It's very symmetrical around it.
If you get rid of useless optional overlaps, you can at least reduce the slope of the trend.
It has a mathematical expectation of 0. That's without the spread-swap-commissions-overlaps. This is without taking into account spread-swap-commissions-overlaps. By the way, because it's 0, you can't roll/slide anything there.
All partly due to the fact that there is no "work on errors", as regulated by the deming cycle (see iso9000 and other MBAs) or the basic management cycle (as in the methods from the USSR).
And why is there none ? because.
How do you calculate this 0? A long time ago I calculated the results of reversalhttps://www.mql5.com/ru/forum/367874/page27#comment_22807159, the average deal size there must be considerably larger than the spread even taking into account the swap which is not charged for the signal. But the experimenter has changed the experiment conditions a hundred times: timeframe, sample size, the Expert Advisor fails, the deer comes, the seal comes. The resulting chart has been compromised and distorted, and it cannot be seriously analyzed.
As for working on errors: since a toy account with instant execution and fixed spread is used, there is no real loss of profit on trades in zeros, because the spread is not widening as it should be.
I've said it again,
If it does trade, why not just test it on just one pair?
over the past year
or on each pair individually,
I don't need 32 pairs, they make everything worse and make trading chaotic,
the only way to get a clear result is to do a test without fills and other human manipulations
If you get rid of the useless optional overlaps, you can at least reduce the slope of the trend.
How did you calculate this 0? A long time ago I estimated the results of reversalhttps://www.mql5.com/ru/forum/367874/page27#comment_22807159. The average deal size must be considerably larger than the spread even taking into account the swap which is not charged for the signal. But the experimenter has changed the experiment conditions a hundred times: timeframe, sample size, the Expert Advisor fails, the deer comes, the seal comes. The resulting chart has been compromised and distorted, and it cannot be seriously analyzed.
As for working on errors: since toy account with instant execution and fixed spread is used, there is no real loss on trades in zeros, the spread is not widening as it should be.
The outfits are wrong. All the line collapse is ensured by the spread and all the variance is by dispersion. MO=0
if PNB suddenly starts "hitting" it means ACF>0.5; i.e. it is actually trading on the past, with a delay. And has positive results only when the past is repeated, in trend.
the fits are wrong. All the line collapse is provided by the spread and all the fluctuations by the variance. MO=0
if PNB suddenly starts to "hit" it means ACF>0.5; i.e. it actually trades on the past, with a delay. And only has positive results when the past is repeated, in a trend.
Why spread? Look at the tab "stats", take the sum of trade results modulo = 138960 points, divide by the total number of deals 1365, we get an average spread of 100 points at 4 signs? Clearly not, there is no such thing.
Why spread? Look at the "stats" tab, take the sum of the results of the trades modulo = 138960 pips, divide by the total number of trades 1365, get an average spread of 100 pips on the 4 digits? Obviously not, there is no such a spread.
If by 4 digits then yes, the spread will have almost no effect.
Why spread? Look at the "stats" tab, take the sum of the results of the trades modulo = 138960 pips, divide by the total number of trades 1365, get an average spread of 100 pips on the 4 digits? Apparently not, it does not happen.
It may happen :-)
I told him a month ago - do not trade at 0:00 there is a huge spread...
And with lots. He opens eurusd, usdchf, eurchf, minus 3 mega spreads at the same time, but there is no use. And he waits for the PNB to tear the ring apart. When there's nothing in the basketurgently to improve arithmetic