Support Resistance Line Break

sadegh

Hi
How can I find days when the resistance line is broken I found local extremum days by this code and I want to find after each extremum when the resistance line was break ! my code:

max_idx = list(argrelextrema(data_f.close.values, np.greater, order=5)[0])
min_idx = list(argrelextrema(data_f.close.values, np.less, order=5)[0])
idx = max_idx + min_idx
idx.sort()

I want to find days like this:
download (1).png
How can I do that with backtrader!

sadegh

How can I find Local extremum? any idea ?

run-out

@sadegh said in Support Resistance Line Break:

argrelextrema
You'll either need to break down the numpy formula and convert that into indicators, or you could use numpy on your data in a dataframe before you enter it into the system, and then just read the signal. You need to make a custom pandas class for your signal. Option two sounds easier.

mics

BT has built in 'Highest' and 'Lowest' indicators that will do this for you.

https://www.backtrader.com/docu/indautoref/#highest
https://www.backtrader.com/docu/indautoref/#lowest

For a long time I used it wrong a received inconsistent results. This structure works for me now:

import backtrader as bt
import backtrader.indicators as bt.ind

class TheStrategy(bt.Strategy):

    params = (
        ('highperiod_0', 60),
        ('lowperiod_0', 60),
        )

def __init__(self):

    self.highest = highest = bt.ind.Highest(self.data, period=self.p.highperiod_0, plot=True, subplot=False)
    self.lowest = lowest = bt.ind.Lowest(self.data, period=self.p.lowperiod_0, plot=True, subplot=False)


def next(self):

    #LONG POSITION: Breakout high when
    self.data.close > self.highest
    self.buy()
    
    #SHORT POSITION: Breakout low when
    self.data.close() < self.lowest
    self.sell

You need to set the period for the number of bars you want to find the highest/lowest data for. Once you have this static high/low working, try using a dynamic indicator for something a little more complex:

https://www.backtrader.com/blog/posts/2018-02-06-dynamic-indicator/dynamic-indicator/