Linear Regression Slope

f1_maik · 2019-09-19T18:40:46+00:00

Hello. I would like to investigate the behavior of the Linear Regression Slope. The indicator is stored in PRT as an oscillator. But what is the code for it? When I have the code, – can I then create a strategy, similar to a kind of Oversold/Overbold strategy, i.e. take the opposite direction at extreme values? – can I put a moving average in the indicator? And can I then create a strategy to open a position in Crosses over/under? – what about divergences? Do I get programmed to find the entry point for divergences or not open a trade if I follow the above strategies? I am sorry. I’m very interested in the indicator, but unfortunately I lack the art of programming here. Maybe one of you can help?

Viewing 4 posts - 1 through 4 (of 4 total)

Author

Posts
09/19/2019 at 7:40 PM #107997 quote

f1_maik
Participant

Junior

Hello.

I would like to investigate the behavior of the Linear Regression Slope.
The indicator is stored in PRT as an oscillator.
But what is the code for it?

When I have the code,
– can I then create a strategy, similar to a kind of Oversold/Overbold strategy, i.e. take the opposite direction at extreme values?
– can I put a moving average in the indicator? And can I then create a strategy to open a position in Crosses over/under?
– what about divergences? Do I get programmed to find the entry point for divergences or not open a trade if I follow the above strategies?

I am sorry. I’m very interested in the indicator, but unfortunately I lack the art of programming here. Maybe one of you can help?

09/19/2019 at 10:24 PM #108014 quote
Vonasi
Moderator

Master
The value returned by the Linear Regression Slope can be easily used in a strategy by using LINEARREGRESSIONSLOPE as shown in the documentation:

LinearRegressionSlope

You can calculate an average of it in your code by just using AVERAGE:
```
 a = average[200](LinearRegressionSlope[10](close))
```
Divergences? Now with that you are entering a whole new kettle of coding fish!
09/20/2019 at 8:02 AM #108038 quote

Nicolas
Keymaster

Legend

f1_maik wrote:

can I then create a strategy, similar to a kind of Oversold/Overbold strategy, i.e. take the opposite direction at extreme values?

Yes, but since the linear regression slope is not normalized, there are many ways to determine what are the oversold and overbought zones such as Bollinger Bands, fixed points distance, highest high / lowest low of X periods, percentile, percentage ..

09/23/2019 at 9:58 PM #108294 quote

f1_maik
Participant

Junior

You dear people. You are so friendly and helpful. Thank you very much. I will try that with the Average. Exactly this program line was missing for me.
Thanks again.
Author

Posts