Common skills for quantitative investment - drawing 3: drawing the golden section line

---

Preface

Previous articles have described how to Draw a line chart of the closing price 、 Candlelight 、 Draw a moving average based on the closing price , Next, how to draw the golden section line

---

You can pay attention to our Tiktok ：“ Financial observation ”（JRGC8888） Learn more about

The golden section

The golden section line is also known to us 0.382,0.618 Split line

Visually 0.382 And visual 0.618

The calculation method is as follows ：

sp382 = (max - min) * 0.382 + min
sp618 = (max - min) * 0.618 + min

Statistically 0.382 And statistical 0.618

We need to use scipy Under the Treasury stats Class scoreatpercentile() Method , Use pip You can install scipy
pip install scipy

How to use it is as follows

• stats.scoreatpercentile(arr, per)
  • arr For input data , It can be a list or DataFrame
  • per The percentage you need to get , The value involved in the calculation is per/100, so 0.628 You should enter per=62.8

sp382_stats = stats.scoreatpercentile(data, 38.2)
sp618_stats = stats.scoreatpercentile(data, 61.8)

Delimit the golden section line

The golden section line delimits two areas 0.382 Split line and two 0.618 The range between , So we use the following function to calculate the range

above618 = np.maximum(sp618, sp618_stats)  #  From vision 0.618 And statistics 0.618 Filter for larger values in 
below618 = np.minimum(sp618, sp618_stats)  #  From vision 0.618 And statistics 0.618 Filter smaller values in 
above382 = np.maximum(sp382, sp382_stats)  #  From vision 0.382 And statistics 0.382 Filter for larger values in 
below382 = np.minimum(sp382, sp382_stats)  #  From vision 0.382 And statistics 0.382 Filter smaller values in 

Draw the golden section

The data we use is through abupy Library imported tsla Historical data of , Combined with the introduction of Draw a candle Methods , The code for drawing the golden section line is as follows

import abupy
import numpy as np
import matplotlib.pyplot as plt
import mpl_finance as mpf
from scipy import stats
from matplotlib.dates import date2num
from abupy import ABuSymbolPd, pd_rolling_mean, nd

# ———————————————————— #
# ————  Default parameter settings  ———— #
# ———————————————————— #
__colorup__ = "red"
__colordown__ = "green"

abupy.env.enable_example_env_ipython()  #  Using sandbox data , The goal is the same data environment as in the book , If not used, an error will be reported 

tsla_df = ABuSymbolPd.make_kl_df('usTSLA', n_folds=2)  #  Fixed import tsla Market data of 
tsla_df = tsla_df[:50]  #  Before selection 50 individual , Too much data makes it difficult to observe 

print(tsla_df[:10])

# ———————————————————— #


def plot_ochl(data_df=tsla_df, axs=None, show=False):
    '''  Draw a candle  :param data_df:  Input data , By default tsla Historical market data of , The input data type currently only supports DataFrame type  :param axs:  Whether to draw on the subgraph  :param show:  Whether to display the image  :return: '''

    drawer = plt if axs is None else axs
    fig, ax = drawer.subplots(figsize=(14, 7))

    qutotes = []

    for index, (d, o, c, h, l) in enumerate(
            zip(data_df.index, data_df.open, data_df.close,
                data_df.high, data_df.low)):
        d = date2num(d)  #  The date of the candle chart should be used matplotlib.finance.date2num To a unique numeric value 
        val = (d, o, c, h, l)  #  date , The opening quotation , The close , The highest , The lowest composition tuple object val
        qutotes.append(val)  #  Add val Join in qutotes

    #  Use mpf.candlestick_ochl Drawing candles ,ochl representative ：open,close,high,low
    mpf.candlestick_ochl(ax, qutotes, width=0.6, colorup=__colorup__, colordown=__colordown__)
    ax.autoscale_view()
    ax.xaxis_date()

    if show:
        plt.show()


def plot_goldline(data_df=tsla_df, axs=None, show=False):
    '''  Draw the golden section  :param data_df:  Input data , By default tsla Historical market data of , The input data type currently only supports DataFrame type  :param axs:  Whether to draw on the subgraph  :param show:  Whether to display the image  :return: '''
    cs_max = data_df.close.max()  #  The maximum value in the closing price series 
    cs_min = data_df.close.min()  #  The lowest value in the closing price series 

    #  Visually 0.382 And visual 0.618
    sp382 = (cs_max - cs_min) * 0.382 + cs_min
    sp618 = (cs_max - cs_min) * 0.618 + cs_min

    #  Statistically 0.382 And statistical 0.618
    sp382_stats = stats.scoreatpercentile(data_df.close, 38.2)
    sp618_stats = stats.scoreatpercentile(data_df.close, 61.8)

    above618 = np.maximum(sp618, sp618_stats)  #  From vision 0.618 And statistics 0.618 Filter for larger values in 
    below618 = np.minimum(sp618, sp618_stats)  #  From vision 0.618 And statistics 0.618 Filter smaller values in 
    above382 = np.maximum(sp382, sp382_stats)  #  From vision 0.382 And statistics 0.382 Filter for larger values in 
    below382 = np.minimum(sp382, sp382_stats)  #  From vision 0.382 And statistics 0.382 Filter smaller values in 

    plt.axhline(sp382, c='r')  #  Horizontal line vision 0.382
    plt.axhline(sp382_stats, c='m')  #  Horizontal line statistics 0.382
    plt.axhline(sp618, c='g')  #  Horizontal line vision 0.618
    plt.axhline(sp618_stats, c='k')  #  Horizontal line statistics 0.618

    plt.fill_between(data_df.index, above618, below618, alpha=0.5, color="r")  #  fill 0.618 red
    plt.fill_between(data_df.index, above382, below382, alpha=0.5, color="g")  #  fill 0.382 green
    plt.legend(['close', 'sp382', 'sp382_stats', 'sp618', 'sp618_stats'], loc='best')  #  Label the names according to the drawing order 

    if show:
        plt.show()


if __name__ == '__main__':
    plot_ochl()  #  Draw a candle , Will create a new canvas 
    plot_goldline(show=True)  #  Draw the golden section 

The output image is as follows ：

Welcome to pay attention to us

Our tiktok ： Financial observation （JRGC8888）