Market phenomena - page 66

 
Can't someone explain it??? It's hard to find anything on the internet about his theory :)
 
But the owner of the coolest PAMM most likely does not think about any market phenomena and guessing on the 4-hour candlesticks and raking in the dough with a shovel.
 
911:
And the owner of the coolest PAMM most likely is not thinking about any market phenomena and guessing at 4-hour candlesticks is earning his money with a shovel.
.
To get to a state where there is "no need to think about it" and "enter the state of complete automatism" if it is manual trading, he most likely had to think a lot and solve a very long series of questions. :)
 
tol64:
To get to a state where you "don't have to think about it" and "get into a state of complete automatism", if it's manual trading, he probably had to think very hard and solve a very long series of questions. :)
Put the context of the market in his mind on a subconscious level?
Although, who knows.... Maybe he's predicting by coffee grounds.
 
Rorschach:

About a year ago I was collecting statistics on the size of H-L candlesticks. The idea was that if in two frames the value of candlesticks exceeds some threshold, there is a high probability of reversal. It did not come to testing but the statistics is still there.

I found probable confirmations of the fact that most participants are oriented at 5m and 1H on the basis of the collected statistics. I built a function on the maxima of probability and fitted it with a power law; the largest errors were at the 5m and 1H.

F-i. The pink one is statistical, the blue one is power law.

Difference between the functions

A lot of work has been done, but what about hypothesis testing? Can it also be done in the echelle?
 
alexeymosc:

What else I will add... Other frames and pairs carry the same ability to work predictably. But I have not checked it.

And another thing - it should not be difficult to make an Expert Advisor (only one adjustable condition, closing of a position by a new bar condition). Maybe I'll make it myself (in my next life), maybe someone will be interested...

So far I've checked some pairs using this strategy "in embryo", mainly majors, with small spreads.

EURUSD M5 works best (it's the one I've started with, my intuition hasn't failed me yet). We trade on the history with the range parameter open[0]-open[1] >= [0.00134] (therefore, the order opens when the price breaks the threshold within the bar). That is all, no more parameters. The spread has been taken as 10 five-figure points. Since March 2011 up to the current date.

The following pairs have performed more or less well : USDJPY, EURJPY, USDCAD. All of them on TF M5.

If you are interested, I will send you excel in my personal message.

 
alexeymosc:

I've checked several pairs using this "in embryo" strategy so far, mostly majors, with small spreads.

EURUSD M5 works best of all (that's what I've started with, my intuition hasn't failed me yet). We trade on the history with the range parameter open[0]-open[1] >= [0.00134] (therefore, the order opens when the price breaks the threshold within the bar). That is all, no more parameters. The spread has been taken as 10 five-figure points. Since March 2011 up to the current date.

The following pairs have performed more or less well: USDJPY, EURJPY, USDCAD. All of them on TF M5.

If you are interested, I will send you excel in my personal message.

That is, setting pending orders to open a bar at a certain distance. When one of them triggers, the other is removed. The position is open. There are no stops, takeovers, and no pending orders either. The position is held up to the next bar. A new cycle starts. Right?
 
tol64:
That is, setting pending orders at the opening of a bar at a certain distance. When one of them triggers, the other one is removed. The position is open. There are no stops, takeovers, and no pending orders either. The position is held up to the next bar. A new cycle starts. Is it so?

That is, setting pending orders to open a bar at a certain distance - YES.

When one is triggered the other is removed. - Not quite. If open[0]-open[1] exceeds the threshold (the only configurable parameter), ONE pending order is placed. If open[0]-open[1] was less than zero, the order is placed on Buy, if the threshold is broken down. If open[0]-open[1] was above zero, an order is placed to sell when the threshold is broken upwards.

No stops, no takeaways, the position is held up until the next bar. Start of a new cycle. Right? - YES.

This whole system is based on a statistical pattern. I introduced the op-open threshold myself, of course, but the system doesn't work without it.

 
Now I'm downloading a 5 minute history with Ducas over 5 years. This is going to be the bomb, I'm going to test the survivability of the strategy on this huge array.
 
alexeymosc:

When one is triggered, the other is removed. - Not quite. If open[0]-open[1] modulo exceeds the threshold (the only configurable parameter), ONE pending order is placed. If open[0]-open[1] was less than zero, the order is placed on Buy, if the threshold is broken down. If open[0]-open[1] was greater than zero, the order is placed to sell on the upwards break of the threshold.

There should be one more parameter. The distance in points at which the order is set. Although it might be the closest possible price. Stop Level + fixed value (pips).