This gist is for another medium article and is about an investment simulator.

simulator.py

import pandas as pd
import numpy as np
import datetime as dt
import math
import warnings

warnings.filterwarnings("ignore")

prices = pd.read_csv("adjclose.csv", index_col="Date", parse_dates=True)
volumechanges = pd.read_csv("volume.csv", index_col="Date", parse_dates=True).pct_change()*100

today = dt.date(2000, 1, 15)
simend = dt.date(2019, 12, 31)
tickers = []
transactionid = 0
money = 1000000
portfolio = {}
activelog = []
transactionlog = []


def getprice(date, ticker):
    global prices
    return prices.loc[date][ticker]


def transaction(id, ticker, amount, price, type, info):
    global transactionid
    if type == "buy":
        exp_date = today + dt.timedelta(days=14)
        transactionid += 1
    else:
        exp_date = today
    if type == "sell":
        data = {"id": id, "ticker": ticker, "amount": amount, "price": price, "date": today, "type": type,
                "exp_date": exp_date, "info": info}
    elif type == "buy":
        data = {"id": transactionid, "ticker": ticker, "amount": amount, "price": price, "date": today, "type": type,
                "exp_date": exp_date, "info": info}
        activelog.append(data)
    transactionlog.append(data)


def buy(interestlst, allocated_money):
    global money, portfolio
    for item in interestlst:
        price = getprice(today, item)
        if not np.isnan(price):
            quantity = math.floor(allocated_money/price)
            money -= quantity*price
            portfolio[item] += quantity
            transaction(0, item, quantity, price, "buy", "")


def sell():
    global money, portfolio, prices, today
    itemstoremove = []
    for i in range(len(activelog)):
        log = activelog[i]
        if log["exp_date"] <= today and log["type"] == "buy":
            tickprice = getprice(today, log["ticker"])
            if not np.isnan(tickprice):
                money += log["amount"]*tickprice
                portfolio[log["ticker"]] -= log["amount"]
                transaction(log["id"], log["ticker"], log["amount"], tickprice, "sell", log["info"])
                itemstoremove.append(i)
            else:
                log["exp_date"] += dt.timedelta(days=1)
    itemstoremove.reverse()
    for elem in itemstoremove:
        activelog.remove(activelog[elem])


def simulation():
    global today, volumechanges, money
    start_date = today - dt.timedelta(days=14)
    series = volumechanges.loc[start_date:today].mean()
    interestlst = series[series > 100].index.tolist()
    sell()
    if len(interestlst) > 0:
        #moneyToAllocate = 500000/len(interestlst)
        moneyToAllocate = currentvalue()/(2*len(interestlst))
        buy(interestlst, moneyToAllocate)


def getindices():
    global tickers
    f = open("symbols.txt", "r")
    for line in f:
        tickers.append(line.strip())
    f.close()


def tradingday():
    global prices, today
    return np.datetime64(today) in list(prices.index.values)


def currentvalue():
    global money, portfolio, today, prices
    value = money
    for ticker in tickers:
        tickprice = getprice(today, ticker)
        if not np.isnan(tickprice):
            value += portfolio[ticker]*tickprice
    return int(value*100)/100


def main():
    global today
    getindices()
    for ticker in tickers:
        portfolio[ticker] = 0
    while today < simend:
        while not tradingday():
            today += dt.timedelta(days=1)
        simulation()
        currentpvalue = currentvalue()
        print(currentpvalue, today)
        today += dt.timedelta(days=7)

main()