A trend-following strategy. - page 2

New comment
 
JLY:
See above: you need a graphical visualisation in the form of an indicator.
The indicator I gave (see link in CodeBase). And if you had told me on which bars you build trend lines and suppo/resistence lines - then I would have made your daily chart on that indicator. Because all levels/lines are plotted by bar/timeframe. By the way, they are dynamic, i.e. may change with every bar or tick for example ... If you do it manually, it would be very troublesome to draw it every time.
 
newdigital:
I gave you the indicator (see link in CodeBase). And if you had told me - on which bars you build trend lines and suppo/resistence lines - then I would have made your daily chart on that indicator. Because all levels/lines are plotted by bar/timeframe. By the way, they are dynamic, i.e. may change with every bar or tick for example ... If they are drawn manually, it's too troublesome to do it all over again.
I wrote above how the lines are drawn ...
 
JLY:

That "algorithm is not for a robot" is what I was quoted from the link)

I think that before writing a robot trader you need to write a base and not to race a demo account. The base is drawing the trend line and its endpoint and identifying the flat extremums. And only then buy and sell signals, Take Profit and Stop Loss. For the manual trading algorithm is used together with drawing trend lines and flat extremums. You see, mathematically it is difficult to calculate, we need a graphic visualization in the form of an indikator.

We need points to plot the trend line and we need points of flat extrema. Three trend lines: flat trend, regular trend and rebound trend.

We take points, e.g. a fractal. To draw a regular trend line, we need a flat breakdown signal.

In the picture: blue line - regular trend line, red line - flat trend, yellow line - breakout trend. The vertical lines show the end of the trend.

When drawing a line naturally the price after the second point must not touch the trend line. After, for a flat trend line this touching may be a signal of "widening of the flat trend line".

A rebound trend line is drawn on the basis of 2 points. If the third maximum or minimum point is found, it will be the flat's opposite extremum. If the opposite extremum is broken, the usual trend line will be drawn. And if the rebound signal at the second point is not confirmed, it will be a breakout signal that will be used to draw a regular trend line.

It goes something like this)


I am not a programmer, but if a programmer understands how to make an indicator according to your description (quote above), then fine. I as a trader cannot trade by this description. It's not a strategy, it's an idea.

 

Question for the topicstarter: What is the main purpose of publishing the strategy? Not judging by any means, just curious about the motivation. Several options come to mind:

a) You want to automate the strategy even though you yourself said it's not for robots;

b) You want the support and approval of your colleagues (perhaps to believe in the system yourself);

c) There is something that bothers you about the system, but you don't know what it is and want help or advice on how to improve it.

d) For the sake of communication.

The question is to see if it makes sense to give advice.

 

JLY:

You need points to draw a trend line, and you need points of flat extremes. Three trend lines: flat trend, regular trend and rebound trend.

We take points, for example a fractal. To draw a regular trend line, we need a breakdown signal of the flat.


And you should give these points to traders so that they trade your strategy normally. On any chart (which corresponds to the strategy) - how points are built (by bar). And the MT-statment - as it is a strategy. And the strategy is something that has already been tried and traded (i.e. - you need an MT report).

 

JLY:

... red is a flat trend ...

Like this)

What colour is the flat trend?

We should also think of the colours for flat flat and trend.

The strategy is otherwise incomprehensible and unalgorithmizable.

 
newdigital:


I as a trader cannot trade according to this description. It's not a strategy, it's an idea.

Well, go for it) I can trade
 
PunkBASSter:

Question for the topicstarter: What is the main purpose of publishing the strategy? Not judging by any means, just curious about the motivation. Several options come to mind:

a) You want to automate the strategy even though you yourself said it's not for robots;

b) You want the support and approval of your colleagues (perhaps to believe in the system yourself);

c) There is something that bothers you about the system, but you don't know what it is and want help or advice on how to improve it.

d) For the sake of communication.

The question is to see if it makes sense to give advice.

for the sake of communication
 
Reshetov:

What colour is the trend flat?

You also need to come up with colouring colours for flat-flat and trend-trend.

Otherwise, the strategy is not fucking understandable and not algorithmic.

There is no "trend flat", "trend flat" or "trend trend flat" in the list of concepts. But in principle there are trends and flats in a trend, and there can be trends and flats in flats.

Brain explosion is to understand a manual algorithm, and about the robot, it is a double explosion).

 
JLY:
for the sake of communication.
If you can trade profitably, good for you. But what you have described is not a strategy, but three strategies in one. A kind of serpent gorynych. It would be much clearer if each strategy was described separately. But it's all just muddled.
123456789...13
New comment