What is financial
trading
FINANCIAL TRADING IN PYTHON
Chelsea Yang
Data Science Instructor
The concept of financial trading
Financial trading is the buying and selling of nancial assets
Various nancial instruments to trade:
Equities
Bonds
Forex
Commodities
Cryptocurrencies
FINANCIAL TRADING IN PYTHON
Why people trade
To make a pro t by taking calculated risks
Long positions: pro t from upward price movement
Short positions: pro t from downward price movement
Market participants:
Institutional traders
Retail traders
FINANCIAL TRADING IN PYTHON
Trading vs. investing
Trading: Investing:
Shorter holding period Longer holding period
Focus on short-term trends or price Focus on market fundamentals
uctuations
Take mostly long positions
Take both long and short positions
FINANCIAL TRADING IN PYTHON
Financial time series data
Time series data: a sequence of data points indexed in time order
import pandas as pd
# Load the data
bitcoin_data = pd.read_csv("bitcoin_data.csv", index_col='Date', parse_dates=True)
print(bitcoin_data.head())
Open High Low Close Volume
Date
2019-01-01 3746.71 3850.91 3707.23 3843.52 4324200990
2019-01-02 3849.22 3947.98 3817.41 3943.41 5244856835
2019-01-03 3931.05 3935.69 3826.22 3836.74 4530215218
2019-01-04 3832.04 3865.93 3783.85 3857.72 4847965467
2019-01-05 3851.97 3904.90 3836.90 3845.19 5137609823
FINANCIAL TRADING IN PYTHON
Plot line chart of time series data
import matplotlib.pyplot as plt
plt.plot(bitcoin_data['Close'], color='red')
plt.title("Daily close price")
plt.show()
FINANCIAL TRADING IN PYTHON
Candlestick chart
Each candlestick displays high, low, open, and close
The color indicates bullish (rising prices) or bearish (falling prices) movement
FINANCIAL TRADING IN PYTHON
Plot candlestick chart with Python
import plotly.graph_objects as go
# Define the candlestick
candlestick = go.Candlestick(
x = bitcoin_data.index,
open = bitcoin_data['Open'],
high = bitcoin_data['High'],
low = bitcoin_data['Low'],
close = bitcoin_data['Close'])
# Create a plot
fig = go.Figure(data=[candlestick])
# Show the plot
fig.show()
FINANCIAL TRADING IN PYTHON
Let's practice!
FINANCIAL TRADING IN PYTHON
Getting familiar with
your trading data
FINANCIAL TRADING IN PYTHON
Chelsea Yang
Data Science Instructor
Different types of traders
Day Trader: holds positions throughout the day but usually not overnight
Swing Trader: holds positions from a few days to several weeks
Position Trader: holds positions from a few months to several years
FINANCIAL TRADING IN PYTHON
Resample the data
Date Close # Resample from hourly to weekly
2019-11-29 04:00:00 1.1010 eurusd_weekly = eurusd_h.resample('W').mean()
2019-11-29 08:00:00 1.1005
2019-11-29 12:00:00 1.0993
Date Close
2019-11-29 16:00:00 1.1016
2019-11-03 1.11248
2019-11-29 20:00:00 1.1020
2019-11-10 1.10860
2019-11-17 1.10208
# Resample from hourly to daily 2019-11-24 1.10659
eurusd_daily = eurusd_h.resample('D').mean() 2019-12-01 1.10113
Date Close
2019-11-25 1.10165
2019-11-26 1.10165
2019-11-27 1.10058
2019-11-28 1.10083
2019-11-29 1.10093
FINANCIAL TRADING IN PYTHON
Calculate daily returns
# Calculate daily returns # Plot the data
stock_data['daily_return'] plt.plot(stock_data['daily_return'])
= stock_data['Close'].pct_change() * 100 plt.show()
Close daily_return
Date
2020-12-11 609.99 -2.723779
2020-12-14 639.83 4.891883
2020-12-15 633.25 -1.028398
2020-12-16 622.77 -1.654955
2020-12-17 655.90 5.319781
FINANCIAL TRADING IN PYTHON
Plot a histogram of daily returns
stock_data['daily_return'].hist(bins=100)
plt.show()
FINANCIAL TRADING IN PYTHON
Data transformation
Technical indicators: various types of data transformations
Simple moving average (SMA): arithmetic mean price over a speci ed n-period
stock_data['sma_50'] = stock_data['Close'].rolling(window=50).mean()
Close sma_50
Date
2020-12-11 122.41 117.7474
2020-12-14 121.78 117.9226
2020-12-15 127.88 118.1502
2020-12-16 127.81 118.4432
2020-12-17 128.70 118.7156
FINANCIAL TRADING IN PYTHON
Plot the rolling average
import matplotlib.pyplot as plt
plt.plot(stock_data['Close'],
label='Close')
plt.plot(stock_data['sma_50'],
label='SMA_50')
plt.legend()
plt.show()
FINANCIAL TRADING IN PYTHON
Let's practice!
FINANCIAL TRADING IN PYTHON
Financial trading
with bt
FINANCIAL TRADING IN PYTHON
Chelsea Yang
Data Science Instructor
The bt package
A exible framework for de ning and backtesting trading strategies
Strategy: a method of buying and selling nancial assets based on prede ned rules
Strategy backtesting: a way to assess the e ectiveness of a strategy by testing it on
historical data
import bt
FINANCIAL TRADING IN PYTHON
The bt process
Step 1: Get the historical price data
Step 2: De ne the strategy
Step 3: Backtest the strategy with the data
Step 4: Evaluate the result
FINANCIAL TRADING IN PYTHON
Get the data
# Download historical prices
bt_data = bt.get('goog, amzn, tsla',
start='2020-6-1', end='2020-12-1')
print(bt_data.head())
goog amzn tsla
Date
2020-06-01 1431.819946 2471.040039 179.619995
2020-06-02 1439.219971 2472.409912 176.311996
2020-06-03 1436.380005 2478.399902 176.591995
2020-06-04 1412.180054 2460.600098 172.876007
2020-06-05 1438.390015 2483.000000 177.132004
FINANCIAL TRADING IN PYTHON
Define the strategy
# Define the strategy
bt_strategy = bt.Strategy('Trade_Weekly',
[bt.algos.RunWeekly(), # Run weekly
bt.algos.SelectAll(), # Use all data
bt.algos.WeighEqually(), # Maintain equal weights
bt.algos.Rebalance()]) # Rebalance
FINANCIAL TRADING IN PYTHON
Backtest
# Create a backtest
bt_test = bt.Backtest(bt_strategy, bt_data)
# Run the backtest
bt_res = bt.run(bt_test)
FINANCIAL TRADING IN PYTHON
Evaluate the result
# Plot the result # Get trade details
bt_res.plot(title="Backtest result") bt_res.get_transactions()
FINANCIAL TRADING IN PYTHON
Let's practice!
FINANCIAL TRADING IN PYTHON
