In one sentence this project is about developing a dashboard connected to the Bittrex exchange API that allows real-time technical analysis indicators to be analysed for all the crptocurrencies traded on Bittrex.
Since May 2017, I have been posting daily technical analysis updates on some of the top cryptocurrencies including Steem where indicators such as the MACD and RSI have been analysed using data from CoinMarketCap. I've received generally positive feedback on the reports but I want the analysis better, more up-to-date and flexible. The best way to do this is to have a real-time connection with the exchanges and have fully up-to-date graphs available 24/7 to interested readers.
In addition to developing a dashboard, I want the platform to serve as a database which will allow analysts to extract detailed data on trade history for each crypotocurrency.
I have already kicked the project off and to date I have created a database and a basic dashboard. The database has been populating since the 23rd January and includes the trade history of 25 crytpocurrencies from the Bittrex API. The data currently includes:
The screenshots below show the dashboard and the structure of the data. You will see that I've included an option allowing the data to be downloaded to CSV.
You can login in below and enter the login details:
username: Matt
password: m@tt!
to select COINS and download the data to a csv.
Note; the data has only been populating since the 23rd January and is limited to the following cryptocurrencies:
# User-defined configuration
coins_list =\
['TIX',
'MER',
'ADT',
'MTL',
'GRS',
'ION',
'VOX',
'WINGS',
'RISE',
'RCN',
'FCT',
'GNT',
'SALT',
'XCP',
'FUN',
'ENG',
'VRC',
'BNT',
'MONA',
'BSD',
'MEME',
'DGD',
'EDG',
'VTC',
'GUP',
'STEEM',
'SBD',
'DNT',
'GNO',
'BLK']
The data being pulled is the trade history from Bittrex (see highlighted area in screenshot below).
The data is fully up-to-date and as granular as you can get. This is very important if you want to analyse market trends for the purposes of trading or just keeping informed on how cryptocurrency prices are trending.
The code is written in Python and the database and dashboard sit in Google Cloud Platform.
All of the code is pushed to my github repository, brophymj/bitgeek.
An extract from the Github code is shown below:
#!/usr/bin/python3
"""Scrapper for cryptocoins historical data."""
import csv
import logging
import traceback
from datetime import datetime
from time import sleep
import requests
from pymongo import MongoClient
# Logger configuration
logging.basicConfig(format='[%(asctime)s] %(levelname)s: %(message)s',
level=logging.INFO, datefmt='%Y/%m/%dT%H:%M:%S')
# MongoDB client configuration
client = MongoClient()
db = client.bittrex
collection = db.market
The database currently pulls trade history data on 25 cryptocurrencies (I plan to expand that to all crypocurrencies traded on Bittrex)
The code connecting to the Bittrex API is shown below:
def write():
"""Create requests to the API and write data."""
for c in coins_list:
paired = 'BTC-{}'.format(c)
logging.info('Parsing [{}]'.format(c))
try:
ops = [{**o, 'Pair': paired} for o in requests.get(
"https://bittrex.com/api/v1.1/public/" +
"getmarkethistory?market={}".format(paired)
).json()['result']
]
logging.info('\tUpdating DB...')
[collection.update_one(
filter={'Id': i['Id'], 'Pair': i['Pair']},
update={'$set': i},
upsert=True
) for i in ops
]
logging.info('\tDone!')
except:
traceback.print_exc()
logging.warning('\tError parsing [{}]!'.format(c))
sleep(1)
Now that I have a database up and running and a very basic dashboard my next steps are as follows:
I want to create algorithms for popular technical analysis indicators like the MACD (Moving Average Convergence Divergence), the RSI (Relative Strength Index), Bollinger Bands and some Volume indicators. These will will be overlaid on the latest datapoints, however, it will also be possible to select a historic period and look at how predictive the indicators were using past data.
To see what these indicators look like in a graph, you can view a recent post where I analysed four such indicators.
https://steemit.com/steem/@bitgeek/steem-price-forecast---27th-january
The graph below shows the MACD doe Steem.
This project is intended to bring this information on a real-time basis and provide customization to enable users to get the most out of the analysis. For example. you will be able to tweak parameters, hide and show trend lines etc.
Another feature I want to include is a display showing daily/weekly return assuming the Buy/Sell signals adopted.
I expect this phase to be complete in the next 4-6 weeks.
The data will initially be confined to the Bittrex API, however, I plan to expand to include other exchange data once phase 1 is complete.
I expect this phase to be complete in two weeks after phase 1.
I'm really excited about this feature. I want to include a notification which will email users when a buy or sell signal is triggered for a particular indicator. Users will be able to specify parameters i.e. magnitude of signal etc. and the notification will include information around expected returns based on similar previous signals, return period etc.
I would really appreciate some feedback and thoughts on the project plan. You can reply in the comment section below, message me on Steemit Chat. The more people involved in this project, the better.
Dashboard picture source: https://pixabay.com/