import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import datetime as dt

fn = 'F-F_Research_Data_5_Factors_2x3.CSV'
FF = pd.read_csv(fn, skiprows=3, nrows = 659)
FF = FF.rename(columns={"Unnamed: 0": "Date"})
FF.index = pd.to_datetime(FF.Date, format = '%Y%m').dt.to_period('m')
FF = FF.drop(['Date'], axis=1)
FF = FF*1e-2

fn = 'DJI.csv'
dji = pd.read_csv(fn)
dji = dji.drop(['Open', 'High', 'Low', 'Adj Close', 'Volume'], axis=1)
dji.index = pd.to_datetime(dji.Date)
dji = dji.resample('M').last()
dji = dji.drop(['Date'], axis=1)
ri = dji.pct_change()[1:]
ri.index = ri.index.to_period('m')
ri = ri.rename(columns={'Close': 'Return'})


# Align monthly data
df = pd.concat([ri, FF], axis=1)
df = df.dropna()