CRYPTO: Getting Started with Binance

+--------------------------------------------+
|              Binance Platform              |
|                                            |
|  +----------+  +----------+  +----------+ |
|  |  Spot    |  | Futures  |  |  Earn    | |
|  | Trading  |  | Trading  |  | (Stake)  | |
|  +----------+  +----------+  +----------+ |
|                                            |
|  +----------+  +----------+  +----------+ |
|  |  REST    |  | WebSocket|  |   Data   | |
|  |   API    |  |  Stream  |  | Analytics| |
|  +----------+  +----------+  +----------+ |
+--------------------------------------------+

pip install python-binance
pip install pandas

from binance import Client
import pandas as pd

# Credentials from your Binance account API management page
API_KEY    = 'your_api_key_here'
API_SECRET = 'your_api_secret_here'

# Initialize the Binance client
client = Client(API_KEY, API_SECRET)

# Verify the connection by fetching account information
account = client.get_account()
print(account)

# Fetch 1-minute candles for BTC/USDT for the last 30 minutes
data = client.get_historical_klines('BTCUSDT', '1m', '30 min ago UTC')

Raw Kline Entry Structure

Field 0:  Open time       (Unix timestamp in milliseconds)
Field 1:  Open price      (string)
Field 2:  High price      (string)
Field 3:  Low price       (string)
Field 4:  Close price     (string)
Field 5:  Volume          (string)
Field 6:  Close time      (Unix timestamp)
Field 7:  Quote volume    (string)
Field 8:  Number of trades
Field 9:  Taker buy base volume
Field 10: Taker buy quote volume
Field 11: Unused field

def gethistory(symbol, interval, lookback):
    """
    Fetch historical OHLCV data from Binance.
    
    Parameters:
        symbol   : Trading pair (e.g., 'BTCUSDT')
        interval : Candle interval (e.g., '1m', '1h')
        lookback : How far back to fetch (e.g., '30 min ago UTC', '1 day ago UTC')
    
    Returns:
        pandas DataFrame with Time index and OHLCV columns
    """
    # Fetch raw data and create DataFrame
    frame = pd.DataFrame(client.get_historical_klines(symbol, interval, lookback))

    # Keep only the first 6 columns: Time, Open, High, Low, Close, Volume
    frame = frame.iloc[:, :6]

    # Assign readable column names
    frame.columns = ['Time', 'Open', 'High', 'Low', 'Close', 'Volume']

    # Set Time as the index
    frame = frame.set_index('Time')

    # Convert Unix milliseconds to human-readable datetime
    frame.index = pd.to_datetime(frame.index, unit='ms')

    # Convert all price/volume strings to float for calculations
    frame = frame.astype(float)

    return frame

# Fetch last 30 minutes of BTC and DOGE data
btc  = gethistory('BTCUSDT',  '1m', '30 min ago UTC')
doge = gethistory('DOGEUSDT', '1m', '30 min ago UTC')

# Visualize the opening price
doge.Open.plot(title='DOGE Open Price - Last 30 Minutes')

# Inspect the data
print(btc.tail())
print(btc.describe())

order = client.create_order(
    symbol='BTCUSDT',
    side='BUY',
    type='MARKET',
    quantity=0.001  # Amount of BTC to buy
)
print(order)

order = client.create_order(
    symbol='BTCUSDT',
    side='SELL',
    type='MARKET',
    quantity=0.001  # Amount of BTC to sell
)
print(order)

Order Response Fields

transactTime   : When the order was executed (Unix timestamp)
orderId        : Unique order identifier
symbol         : Trading pair
side           : BUY or SELL
type           : MARKET or LIMIT
status         : FILLED, PARTIALLY_FILLED, etc.
executedQty    : How much was actually filled
fills          : List of individual fill details (price, qty, commission)

# Calculate how the asset performed over the lookback period
# pct_change(): percentage change between each row
# +1: converts percentage to growth factor (e.g., -0.02 becomes 0.98)
# cumprod(): compounds the growth factors across all rows
# -1: converts final compounded factor back to net return

cumulret = (df.Open.pct_change() + 1).cumprod() - 1

# Example: if cumulret[-1] is -0.003, the asset fell 0.3% in the period