What is the meaning of the term non-lagging (in relation to the indicator)? - page 4

 
Женя:

Really? Can you show me on the SB? On the sine wave, I can do it too))

I'm lazy on the SB, but I can show you on the history. Tomorrow night, probably better in person, so I can get it. Just remind me.
 
Aleksey Vyazmikin:

The results in MO can vary, even above 55% and for some strategies 50% is very good - trend strategies for example.

But I'm talking about the qualitatively different approach - when we do not wait for the indicator readings but make a decision a couple of steps ahead knowing what the indicator will show, if the price moves in this or that direction. This approach allows us to consider the possibility of entering a position now, if we know. The indicator will most likely bounce off the RSI, and stops can be set very low.

How do we know this? Only by predicting with MO by picking a target and chips. Unfortunately, no matter what kind of "popular data" we use, whatever targets and features we invent, the figures are almost the same, if we don't cheat ourselves, and this will in any case come out during backtesting.

 
Женя:

How can we know this? Also, we can only predict with the help of MO by selecting a target and features. Alas, no matter how twisted you get with "folk data", whatever targets and features you invent, from those that can be traded, the numbers are about the same, unless you cheat yourself, which in any case will come out in the backtesting.

Years of searching and sticking to the monitor can only give it away.

Here is the forecast, I am in because I expect the price will most likely reach RSI (70) and (or) the daily ATR 100%, from there the probability of correction on M1 (I trade on M1, but the scale shows better on M15) will increase.

Screenshots from MetaTrader platform

Si-6.19, M15, 2019.03.22

JSC ''Otkritie Broker'', MetaTrader 5, Real

RSI lvl_70 H1 & ATR D1 100% reversal forecast

Si-6.19, M15, 2019.03.22, Otkritie Broker, MetaTrader 5, Real


 
Aleksey Vyazmikin:

Years of searching and staring at the monitor can only give that.

Here is the forecast, I enter because I expect the price will most likely reach RSI (70) and (or) the daily ATR 100%, the probability of a correction on M1 will increase (I trade on M1, but the scale shows better on M15).

I know what you mean, I used to be like that, but then I realized that "there is no Father Christmas", it hurts, but it's true(

Although condemn for this type of pareidolic gambling I would not dare, from my own experience I know that it's sometimes all very "real", especially when a certain time lucky ...
 
Женя:

Yes, I know what you mean, I used to be like that myself, but then I realised that 'there is no Father Christmas', it hurts but it's true((

You just have to have faith...

Here is the result - the corrective segment was drawn from three significant resistances - RSI 70, ATR 100 and Fibonacci levels 0% and also evening of March 19 - these levels are often a bounce.


 
Aleksey Vyazmikin:

You just have to have faith...

Here's the result - the corrective stretch is drawn from three significant resistances - RSI 70, ATR 100 and Fibonacci 0% level.

I'm the same as you so far, in fact...

It's just fun, the main thing is not to risk significant sums :)

 
khorosh:

So that there is no ambiguity about which lag we are talking about.


Colour change - the sell signal appeared 5 bars late in relation to the bar where the zigzag extremum was.

There is nothing wrong with this lag. There are no beautiful reversals and MARK gives time to enter later on the pullback.
That's probably the right thing to do.

 
Женя:

I'm just like you so far, in fact...

It's just fun, the main thing is not to risk substantial amounts of money :)

It's the kind of nuances you don't know how to teach MO models - you seem to tell them everything, give them a piece of cake, make a bunch of predictors, but they don't get the gist of the idea.

The idea of manual trading is a lot of unnecessary emotion, which usually does not end well.

 
Aleksey Vyazmikin:

The main thing is that you do not know how to teach the MoD models - you tell them everything, you explain it to them, you make a lot of predictors, but they do not get the idea.

I have been teaching the MoD with my data, but I have not been able to get them to understand it.

You have trained the MoD on my data, and it even seemed to work out well. And there were no predictors there, only a normalized price series.

It is important what to teach and predictors will be thought up by MI itself. ) If it is possible in principle. And if not, no predictors will help.

 
Yuriy Asaulenko:

You trained MoD on my data, and it seemed to work quite well. And there were no predictors, only a normalized price series.

The important thing is what to teach, and the predictors will come up on their own.

I've definitely worked with the data, but I don't have a deep understanding of how to get it right and make a strategy out of it.