AI & Machine Learning Trading Strategies:Complete Guide 2025

import yfinance as yf
import pandas as pd

# Download EURUSD data
data = yf.download('EUR=X', start='2020-01-01', end='2024-12-31')
data.to_csv('eurusd_data.csv')
print(f"Downloaded {len(data)} records")

# Step 2: Create Features
import pandas as pd

# Create moving averages
data['MA_10'] = data['Close'].rolling(window=10).mean()
data['MA_20'] = data['Close'].rolling(window=20).mean()
data['MA_50'] = data['Close'].rolling(window=50).mean()

# Create target variable (what we want to predict)
data['Future_Return'] = (data['Close'].shift(-1) - data['Close']) / data['Close']

print("✓ Features created: MA_10, MA_20, MA_50, Future_Return")

# Step 3: Prepare Data
from sklearn.model_selection import train_test_split

# Select features
features = ['MA_10', 'MA_20', 'MA_50']
X = data[features].dropna()  # Features (inputs)
y = data['Future_Return'].loc[X.index]  # Target (what to predict)

# Split into training (80%) and testing (20%)
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, shuffle=False
)

print(f"✓ Training set: {len(X_train)} samples")
print(f"✓ Test set: {len(X_test)} samples")

# Step 4: Train Model
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error

# Create and train the model
model = LinearRegression()
model.fit(X_train, y_train)  # Learn from training data

# Make predictions on test data
predictions = model.predict(X_test)

# Calculate error
mse = mean_squared_error(y_test, predictions)
print(f"✓ Model trained! Mean Squared Error: {mse:.6f}")

# Step 5: Generate Signals
# Convert predictions to trading signals
signals = []
for p in predictions:
    if p > 0.001:  # Predicted return > 0.1%
        signals.append('BUY')
    elif p < -0.001:  # Predicted return < -0.1%
        signals.append('SELL')
    else:
        signals.append('HOLD')

# Add signals to test data
test_data = X_test.copy()
test_data['Signal'] = signals
test_data['Predicted_Return'] = predictions

print(f"✓ Generated {signals.count('BUY')} BUY signals")
print(f"✓ Generated {signals.count('SELL')} SELL signals")

# Step 6: Backtest
initial_balance = 10000
balance = initial_balance
position = None  # 'LONG', 'SHORT', or None
entry_price = 0

for i, row in test_data.iterrows():
    signal = row['Signal']
    current_price = data.loc[i, 'Close']
    
    # Handle BUY signal
    if signal == 'BUY' and position != 'LONG':
        # Close SHORT position if open
        if position == 'SHORT':
            pnl = (entry_price - current_price) / entry_price * balance
            balance += pnl
        # Open LONG position
        position = 'LONG'
        entry_price = current_price
        
    # Handle SELL signal
    elif signal == 'SELL' and position != 'SHORT':
        # Close LONG position if open
        if position == 'LONG':
            pnl = (current_price - entry_price) / entry_price * balance
            balance += pnl
        # Open SHORT position
        position = 'SHORT'
        entry_price = current_price

# Calculate final results
total_return = (balance - initial_balance) / initial_balance * 100
print(f"✓ Backtest complete!")
print(f"Initial Balance: ${initial_balance:,.2f}")
print(f"Final Balance: ${balance:,.2f}")
print(f"Total Return: {total_return:.2f}%")

# Complete AI Trading System
import yfinance as yf
import pandas as pd
import numpy as np
from sklearn.linear_model import LinearRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error
import matplotlib.pyplot as plt

print("=" * 50)
print("AI Trading System - Starting...")
print("=" * 50)

# Step 1: Download data
print("
[1/6] Downloading historical data...")
data = yf.download('EUR=X', start='2020-01-01', end='2024-12-31', progress=False)
print(f"✓ Downloaded {len(data)} records")

# Step 2: Create features
print("
[2/6] Creating features...")
data['MA_10'] = data['Close'].rolling(window=10).mean()
data['MA_20'] = data['Close'].rolling(window=20).mean()
data['MA_50'] = data['Close'].rolling(window=50).mean()
data['Future_Return'] = (data['Close'].shift(-1) - data['Close']) / data['Close']
print("✓ Features created")

# Step 3: Prepare data
print("
[3/6] Preparing data for training...")
features = ['MA_10', 'MA_20', 'MA_50']
X = data[features].dropna()
y = data['Future_Return'].loc[X.index]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, shuffle=False)
print(f"✓ Training set: {len(X_train)} samples, Test set: {len(X_test)} samples")

# Step 4: Train model
print("
[4/6] Training AI model...")
model = LinearRegression()
model.fit(X_train, y_train)
predictions = model.predict(X_test)
mse = mean_squared_error(y_test, predictions)
print(f"✓ Model trained! MSE: {mse:.6f}")

# Step 5: Generate signals
print("
[5/6] Generating trading signals...")
signals = ['BUY' if p {'>'} 0.001 else 'SELL' if p < -0.001 else 'HOLD' for p in predictions]
test_data = X_test.copy()
test_data['Signal'] = signals
test_data['Predicted_Return'] = predictions
print(f"✓ Generated {signals.count('BUY')} BUY, {signals.count('SELL')} SELL signals")

# Step 6: Backtest
print("
[6/6] Running backtest...")
initial_balance = 10000
balance = initial_balance
position = None
entry_price = 0

for i, row in test_data.iterrows():
    signal = row['Signal']
    current_price = data.loc[i, 'Close']
    
    if signal == 'BUY' and position != 'LONG':
        if position == 'SHORT':
            pnl = (entry_price - current_price) / entry_price * balance
            balance += pnl
        position = 'LONG'
        entry_price = current_price
        
    elif signal == 'SELL' and position != 'SHORT':
        if position == 'LONG':
            pnl = (current_price - entry_price) / entry_price * balance
            balance += pnl
        position = 'SHORT'
        entry_price = current_price

total_return = (balance - initial_balance) / initial_balance * 100
print(f"✓ Backtest complete!")

# Results
print("
" + "=" * 50)
print("RESULTS")
print("=" * 50)
print(f"Initial Balance: {{$initial_balance:,.2f}}")
print(f"Final Balance: {{$balance:,.2f}}")
print(f"Total Return: {total_return:.2f}%")
print("=" * 50)