Stock market forecasting : Fundamental analysis

Theoretical report
Experimental report

Preparing the financial data (news & market values)

from FinancialData import financial_data
iex = financial_data()
print(iex.get_trade_bars(['AAPL'], '1m')[:10])
print(iex.get_news(['AAPL', 'MSFT'])[['symbol','time','headline']][:10])

# list(iex.all_ticker()) : list off all valid securities' symbols
# This step could take a while, considering that this API provides 8750 security's related news
df = iex.get_news(list(iex.all_ticker()))
df.to_excel("assets/news.xlsx")

##### IOdy.py : ######
import pandas as pd
from datetime import datetime, timedelta
from pandas_datareader import data
import numpy as np

news = pd.read_excel('assets/news.xlsx', index_col=0)

#remove news that are published earlier than 30 days from today
absolute_thresh = pd.Timestamp(datetime.today().date()-timedelta(days=37))
news = pd.read_excel('assets/news.xlsx', index_col=0)
news = news.loc[news['time']>=absolute_thresh]
news[['symbol', 'time', 'headline', 'summary']].head(2)

	symbol	time	headline	summary
0	KAI	2019-04-04 16:25:54	12 Upcoming Dividend Increases	I'm a huge fan of dividend growth stocks an...
1	ALTM	2019-04-12 21:27:00	FERC clears EPIC pipeline rates, three others ...	The 550K bbl/day EPIC pipeline wins appr...

# This function uses Yahoo Finance to market data
def get_market(security, start_date, end_date):
	end_date_str = end_date.strftime("%Y-%m-%d")
	start_date_str = start_date.strftime("%Y-%m-%d")
	asset_ = data.DataReader(security,  start=start_date_str, end=end_date_str, data_source='yahoo')
	return asset_

get_market('AAPL', intervals_[0][0], intervals_[0][1])

	High	Low	Open	Close	Volume	Adj Close
Date
2019-03-21	196.330002	189.809998	190.020004	195.089996	51034200	195.089996
2019-03-22	197.690002	190.779999	195.339996	191.050003	42407700	191.050003