Stock/pandas_eval.py

import pandas as pd
import pandas_datareader.data as web
import numpy as np
import math
import matplotlib.pyplot as plt
import datetime as dt
import os

# Stock Investors Financial Math
# https://www.fmlabs.com/reference/default.htm?url=SimpleMA.htm

# Alpha Vantage
# https://www.alphavantage.co/documentation/

# Pandas
# https://pandas.pydata.org/pandas-docs/stable/index.html

# Knowledge
# https://www.investopedia.com/articles/technical/02/050602.asp
# https://ntguardian.wordpress.com/2016/09/19/introduction-stock-market-data-python-1

key = '0UO7Z2MVZ2YSQSVE'

def ema(data, key, alpha=0.5, ic=None):
	result = []
	if ic is None:
		r = data[key][0]
	else:
		r = ic
	for v in data[key]:
		r = alpha*r + (1-alpha)*v
		result.append(r)

	return np.array(result)

def sma(data, key, window_days, ic=None):
	mem = [0] * window_days
	result = []
	k=0
	cumsum = 0
	for v in data[key]:
		cumsum += (v - mem[k])
		mem[k] = v
		k += 1
		if k >= window_days:
			k=0

		result.append(cumsum/window_days)

	return result

def sd(data, key, window_days, ic=None):
	mem = [0] * window_days
	result = []
	k=0
	cumsum = 0
	data_mean = data[key] - sma(data, key, window_days, ic)
	for v in data_mean:
		v2 = v * v
		cumsum += (v2 - mem[k])
		mem[k] = v2
		k += 1
		if k >= window_days:
			k=0

		var = max(0, cumsum) / window_days
		result.append(math.sqrt(var))

	return result


def colsum(data, keys):
	result = np.zeros(data.shape[0])
	for key in keys:
		result += data[key]

	return result

def macd(data, key):
	ema_short = ema(data, key, alpha=0.85)
	ema_long = ema(data, key, alpha=0.925)

	return ema_short - ema_long

def bollinger(data, window_days, f=2):
	tp_ser = colsum(data, keys=['high', 'low', 'close']) / 3
	tp_df = tp_ser.to_frame(name='tp')
	stddev = np.array(sd(tp_df, 'tp', window_days))

	mid = np.array(sma(tp_df, key='tp', window_days=window_days))
	upper = mid + f * stddev
	lower = mid - f * stddev
	result = pd.DataFrame(upper, index=data.index, columns=['upper'])
	result['lower'] = lower
	result['mid'] = mid
	return result

data = {}
titles = []
titles.append('AAPL')
titles.append('XLNX')
titles.append('QCOM')
titles.append('DPW.DE')
titles.append('CSCO')
titles.append('AIR')

show_title = 'QCOM'
show_range_days = 20

pd.DatetimeIndex
def fetch(title):
	filename = title + '.h5'
	fetch = False
	today = dt.date.today()

	try:
		hdf = pd.HDFStore(filename, 'r')
		hdf.close()
	except:
		fetch = True

	try:
		lastmodified = dt.datetime.fromtimestamp(os.stat(filename).st_mtime).date()
		if lastmodified != today:
			fetch = True
	except:
		fetch = True

	if fetch:
		print ("Fetching \"{}\"".format(title))
		# Get stock price via data reader
		start = dt.datetime(2019,1,1)
		end = dt.date.today()
		data = web.DataReader(title, "av-daily", start, end, api_key=key)
		hdf = pd.HDFStore(title + '.h5')
		hdf[title] = data
	else:
		hdf = pd.HDFStore(filename, 'r')
		data = hdf[title]
	hdf.close()

def show(title):

	k_euro = 1/1.11

	filename = title + '.h5'

	hdf = pd.HDFStore(filename, 'r')
	data_full = hdf[title]*k_euro

	data_full['ema'] = ema(data_full, key='close', alpha=0.75)
	data_full['macd'] = macd(data_full, key='close')
	data_full['sma'] = sma(data_full, key='close', window_days=30)
	boll_full = bollinger(data_full, window_days=30)
	hdf.close()

	data = data_full[len(data_full)-1 - show_range_days:]
	boll = boll_full[len(boll_full)-1 - show_range_days:]
	dates = data.index

	fig = plt.figure(1)

	ax = plt.subplot(311)

	def hover(event):
		if event.inaxes == ax:
			x_idx = int(event.xdata)
			ax.format_xdata = lambda x: dates[x_idx]
			ax.format_ydata = lambda y: '{:.2f}'.format(data['close'][x_idx])

	fig.canvas.mpl_connect("motion_notify_event", hover)

	data['close'].plot()
	data['ema'].plot()
	data['sma'].plot()
	plt.title(title)
	plt.legend()
	plt.grid()

	plt.subplot(312)
	boll['upper'].plot()
	boll['mid'].plot()
	boll['lower'].plot()
	plt.legend()
	plt.grid()

	plt.subplot(313)
	data['macd'].plot()
	plt.legend()
	plt.grid()

	plt.show()



for title in titles:
	fetch(title)

show(show_title)