Read in tick data sourced from the `bitcoincharts` api.

read_in_btc_usd_tick_data.py

# Step 1: Download zip file for bitstampUSD 
##### Link: -> http://api.bitcoincharts.com/v1/csv/
##### See SO post for more detail: https://stackoverflow.com/questions/16143266/get-bitcoin-historical-data

# Step 2: Extract CSV file and save in downloads
# Step 3: Run the below code reading and cleaning the data to save as minute ohlc data...

import os
import numpy as np
import pandas as pd

downloads_path = os.path.expanduser('~/Downloads')
btc_data = pd.read_csv(os.path.join(downloads_path, "bitstampUSD.csv"), header=None)
btc_data.columns = ["TimeStamp", "Price", "Volume"]
# Convert unix timestamp to datetime
btc_data['TimeStamp'] = pd.to_datetime(btc_data['TimeStamp'], unit='s')
btc_data.set_index("TimeStamp", inplace=True)
# Filter out Older data
btc_data.loc[btc_data.index >= "2017-04-01"]

btc_data.head(3)

ohlc_data = pd.DataFrame()

# Group by minute and get OHLC -> Warning takes a while to run for the full dataset
grp_btc_data = btc_data.groupby(pd.Grouper(freq='min'))
ohlc_data["volume"] = grp_btc_data["Volume"].sum()
ohlc_data["open"] = grp_btc_data["Price"].nth(0)
ohlc_data["high"] = grp_btc_data["Price"].max()
ohlc_data["low"] = grp_btc_data["Price"].min()
ohlc_data["close"] = grp_btc_data["Price"].nth(-1)