{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Overview\n", "This project focuses on developing a reinforcement learning-based algorithmic trading strategy with the goal of creating a trading agent that learns optimal trading strategies by interacting with historical market data and making buy/sell decisions based on current market conditions. The reinforcement learning-based trading strategy learns to make trading decisions based on historical market data. The Q-learning algorithm is used to learn the optimal actions (buy, hold, or sell) for different states of the market. The agent is trained through iterative episodes, where it explores and exploits different actions and receives rewards based on the profitability of its decisions.\n", "\n", "The performance of the reinforcement learning strategy is compared to a baseline buy-and-hold strategy. The results show that the reinforcement learning strategy outperforms the buy-and-hold strategy in terms of cumulative returns and Sharpe ratio. However, it also experiences a higher maximum drawdown, indicating potential risks. Overall, the project demonstrates the application of reinforcement learning techniques in developing algorithmic trading strategies. It highlights the potential of using machine learning to learn optimal trading strategies from historical data. However, it also acknowledges the complexities and computational requirements associated with reinforcement learning algorithms.\n", "\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Part 1\n", "Collect historical market data for a set of assets (e.g., stocks, cryptocurrencies). Preprocess the data to remove outliers, handle missing values, and format it into a suitable input for the reinforcement learning model." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
AMZNAXPAMGNAAPLBACATCSCOCVXGSHD...NKEPGTRVUNHCRMVZVWMTDISDOW
Date
2022-09-29 00:00:00-04:00114.800003134.829849217.547775141.273102125.330002160.74656738.699997136.735626283.136993267.565887...93.491486123.847023149.782745497.780243146.80999834.812500177.91124043.25796597.13343040.799973
2022-09-30 00:00:00-04:00113.000000132.011749214.680923137.029358121.080002158.98310938.156269135.696655280.211060265.268372...81.516960121.489410148.349609494.072540143.83999634.208866175.53001442.42388594.02356740.587479
2022-10-03 00:00:00-04:00115.879997137.011948219.481247141.243378126.050003165.91098039.386806143.309326286.043854272.728302...83.752991123.664177152.097092504.315186147.89999435.280991179.48226943.34955696.81446841.834751
2022-10-04 00:00:00-04:00121.089996142.335068221.938553144.862442133.509995174.05006440.262924148.881912301.075134278.361664...86.930504125.194252156.357788511.808838155.72999635.866600183.43452543.912155101.11047443.072800
2022-10-05 00:00:00-04:00120.949997141.268494222.700500145.159912132.110001172.83888240.426598149.731979295.462280278.640442...89.343056124.328163155.457230515.624146156.22999635.497215185.43042043.477119100.47254942.555408
\n", "

5 rows × 30 columns

\n", "
" ], "text/plain": [ " AMZN AXP AMGN AAPL \\\n", "Date \n", "2022-09-29 00:00:00-04:00 114.800003 134.829849 217.547775 141.273102 \n", "2022-09-30 00:00:00-04:00 113.000000 132.011749 214.680923 137.029358 \n", "2022-10-03 00:00:00-04:00 115.879997 137.011948 219.481247 141.243378 \n", "2022-10-04 00:00:00-04:00 121.089996 142.335068 221.938553 144.862442 \n", "2022-10-05 00:00:00-04:00 120.949997 141.268494 222.700500 145.159912 \n", "\n", " BA CAT CSCO CVX \\\n", "Date \n", "2022-09-29 00:00:00-04:00 125.330002 160.746567 38.699997 136.735626 \n", "2022-09-30 00:00:00-04:00 121.080002 158.983109 38.156269 135.696655 \n", "2022-10-03 00:00:00-04:00 126.050003 165.910980 39.386806 143.309326 \n", "2022-10-04 00:00:00-04:00 133.509995 174.050064 40.262924 148.881912 \n", "2022-10-05 00:00:00-04:00 132.110001 172.838882 40.426598 149.731979 \n", "\n", " GS HD ... NKE PG \\\n", "Date ... \n", "2022-09-29 00:00:00-04:00 283.136993 267.565887 ... 93.491486 123.847023 \n", "2022-09-30 00:00:00-04:00 280.211060 265.268372 ... 81.516960 121.489410 \n", "2022-10-03 00:00:00-04:00 286.043854 272.728302 ... 83.752991 123.664177 \n", "2022-10-04 00:00:00-04:00 301.075134 278.361664 ... 86.930504 125.194252 \n", "2022-10-05 00:00:00-04:00 295.462280 278.640442 ... 89.343056 124.328163 \n", "\n", " TRV UNH CRM VZ \\\n", "Date \n", "2022-09-29 00:00:00-04:00 149.782745 497.780243 146.809998 34.812500 \n", "2022-09-30 00:00:00-04:00 148.349609 494.072540 143.839996 34.208866 \n", "2022-10-03 00:00:00-04:00 152.097092 504.315186 147.899994 35.280991 \n", "2022-10-04 00:00:00-04:00 156.357788 511.808838 155.729996 35.866600 \n", "2022-10-05 00:00:00-04:00 155.457230 515.624146 156.229996 35.497215 \n", "\n", " V WMT DIS DOW \n", "Date \n", "2022-09-29 00:00:00-04:00 177.911240 43.257965 97.133430 40.799973 \n", "2022-09-30 00:00:00-04:00 175.530014 42.423885 94.023567 40.587479 \n", "2022-10-03 00:00:00-04:00 179.482269 43.349556 96.814468 41.834751 \n", "2022-10-04 00:00:00-04:00 183.434525 43.912155 101.110474 43.072800 \n", "2022-10-05 00:00:00-04:00 185.430420 43.477119 100.472549 42.555408 \n", "\n", "[5 rows x 30 columns]" ] }, "execution_count": 1, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import yfinance as yf\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import warnings\n", "warnings.filterwarnings('ignore')\n", "import pandas as pd\n", "\n", "# List of DJIA ticker symbols\n", "djia_tickers = [\n", " 'AMZN', 'AXP', 'AMGN', 'AAPL', 'BA', 'CAT', 'CSCO', 'CVX', 'GS', 'HD',\n", " 'HON', 'IBM', 'INTC', 'JNJ', 'KO', 'JPM', 'MCD', 'MMM', 'MRK', 'MSFT',\n", " 'NKE', 'PG', 'TRV', 'UNH', 'CRM', 'VZ', 'V', 'WMT', 'DIS', 'DOW'\n", "]\n", "\n", "# Create an empty DataFrame to store closing prices\n", "closing_prices = pd.DataFrame()\n", "\n", "# Fetch closing prices for each ticker symbol\n", "for ticker in djia_tickers:\n", " ticker_data = yf.Ticker(ticker)\n", " ticker_df = ticker_data.history(period='365d')\n", " closing_prices[ticker] = ticker_df['Close']\n", "\n", "# Drop rows with missing data\n", "closing_prices = closing_prices.dropna()\n", "\n", "closing_prices.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Part 2\n", "Implement a reinforcement learning algorithm (e.g., Q-learning, Deep Q-Networks) to learn trading strategies. Design the state space, action space, and reward structure for the trading agent. Train the reinforcement learning model using the preprocessed historical data." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[[ 9.46252188e-02 1.14806924e-01 1.38531837e-01]\n", " [ 1.29176448e-01 9.76606225e-02 7.16049612e-02]\n", " [ 1.15946476e-01 8.28487120e-02 4.84086391e-02]\n", " ...\n", " [-3.61182166e-04 7.75066180e-07 -1.20506786e-03]\n", " [-2.00493522e-04 0.00000000e+00 -1.13303229e-02]\n", " [ 0.00000000e+00 0.00000000e+00 0.00000000e+00]]\n" ] } ], "source": [ "# Define the state space\n", "states = closing_prices.pct_change().dropna().values\n", "\n", "# Define the action space\n", "actions = ['buy', 'hold', 'sell']\n", "\n", "# Define the reward structure\n", "def calculate_reward(action, price_change):\n", " if action == 'buy':\n", " return price_change\n", " elif action == 'sell':\n", " return -price_change\n", " else:\n", " return 0\n", "\n", "# Initialize the Q-table\n", "q_table = np.zeros((len(states), len(actions)))\n", "\n", "# Set hyperparameters\n", "alpha = 0.1 # Learning rate\n", "gamma = 0.9 # Discount factor\n", "epsilon = 0.1 # Exploration rate\n", "\n", "# Training loop\n", "num_episodes = 1000\n", "for episode in range(num_episodes):\n", " state = 0 # Start from the first time step\n", " done = False\n", " \n", " while not done:\n", " # Choose an action using epsilon-greedy policy\n", " if np.random.uniform(0, 1) < epsilon:\n", " action = np.random.choice(actions)\n", " else:\n", " action = actions[np.argmax(q_table[state])]\n", " \n", " # Take the action and observe the next state and reward\n", " stock_index = np.random.randint(0, states.shape[1]) # Randomly select a stock\n", " price_change = states[state, stock_index]\n", " reward = calculate_reward(action, price_change)\n", " \n", " # Update the Q-value\n", " next_state = state + 1 # Move to the next time step\n", " q_table[state, actions.index(action)] += alpha * (\n", " reward + gamma * np.max(q_table[next_state]) - q_table[state, actions.index(action)]\n", " )\n", " \n", " # Move to the next state\n", " state = next_state\n", " \n", " # Check if the episode is done\n", " if state == len(states) - 1:\n", " done = True\n", "\n", "# Print the learned Q-table\n", "print(q_table)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The script uses Q-learning to learn the optimal trading strategy. It initializes a Q-table and iteratively updates it based on the observed states, actions, and rewards. The agent explores the environment using an epsilon-greedy policy, gradually learning the optimal action for each state." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Part 3.1\n", "Evaluate the performance of the trained trading agent using a separate testing dataset. Implement backtesting to assess the profitability and risk-adjusted returns of the trading strategy. " ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Total reward on testing data: 0.07467751259669841\n" ] } ], "source": [ "# Split the data into training and testing sets\n", "train_data = closing_prices.iloc[:int(len(closing_prices) * 0.8)]\n", "test_data = closing_prices.iloc[int(len(closing_prices) * 0.8):]\n", "\n", "# Define the state space for training and testing\n", "train_states = train_data.pct_change().dropna().values\n", "test_states = test_data.pct_change().dropna().values\n", "\n", "# Train the Q-learning agent using the training data\n", "# Initialize the Q-table for training\n", "q_table_train = np.zeros((len(train_states), len(actions)))\n", "\n", "# Set hyperparameters\n", "alpha = 0.1 # Learning rate\n", "gamma = 0.9 # Discount factor\n", "epsilon = 0.1 # Exploration rate\n", "\n", "# Training loop\n", "num_episodes = 1000\n", "for episode in range(num_episodes):\n", " state = 0 # Start from the first time step\n", " done = False\n", " \n", " while not done:\n", " # Choose an action using epsilon-greedy policy\n", " if np.random.uniform(0, 1) < epsilon:\n", " action = np.random.choice(actions)\n", " else:\n", " action = actions[np.argmax(q_table_train[state])]\n", " \n", " # Take the action and observe the next state and reward\n", " stock_index = np.random.randint(0, train_states.shape[1]) # Randomly select a stock\n", " price_change = train_states[state, stock_index]\n", " reward = calculate_reward(action, price_change)\n", " \n", " # Update the Q-value\n", " next_state = state + 1 # Move to the next time step\n", " q_table_train[state, actions.index(action)] += alpha * (\n", " reward + gamma * np.max(q_table_train[next_state]) - q_table_train[state, actions.index(action)]\n", " )\n", " \n", " # Move to the next state\n", " state = next_state\n", " \n", " # Check if the episode is done\n", " if state == len(train_states) - 1:\n", " done = True\n", "\n", "# Evaluate the trained agent on the testing data\n", "state = 0\n", "done = False\n", "total_reward = 0\n", "\n", "while not done:\n", " # Choose the action with the highest Q-value\n", " action = actions[np.argmax(q_table[state])]\n", " \n", " # Take the action and observe the reward\n", " stock_index = np.random.randint(0, test_states.shape[1])\n", " price_change = test_states[state, stock_index]\n", " reward = calculate_reward(action, price_change)\n", " \n", " # Accumulate the rewards\n", " total_reward += reward\n", " \n", " # Move to the next state\n", " state += 1\n", " \n", " # Check if the episode is done\n", " if state == len(test_states) - 1:\n", " done = True\n", "\n", "# Print the total reward obtained on the testing data\n", "print(\"Total reward on testing data:\", total_reward)" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Backtesting Results:\n", " Action Price_Change Reward Portfolio_Value\n", "0 sell 0.007740 -0.007740 9912.604800\n", "1 buy 0.005091 0.005091 9953.159317\n", "2 buy -0.000896 -0.000896 9934.287690\n", "3 hold -0.009465 0.000000 9934.287690\n", "4 sell 0.011137 -0.011137 9813.719394\n", ".. ... ... ... ...\n", "67 buy 0.006580 0.006580 8690.620577\n", "68 hold 0.003262 0.000000 8690.620577\n", "69 hold 0.002609 0.000000 8690.620577\n", "70 buy 0.000409 0.000409 8685.483530\n", "71 buy 0.005291 0.005291 8722.751037\n", "\n", "[72 rows x 4 columns]\n", "\n", "Performance Metrics:\n", "Total Return: -0.1277248963127362\n", "Sharpe Ratio: -3.0731337193606145\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Create a DataFrame to store the backtesting results\n", "backtest_results = pd.DataFrame(columns=['Action', 'Price_Change', 'Reward', 'Portfolio_Value'])\n", "\n", "# Set the initial portfolio value\n", "initial_portfolio_value = 10000\n", "portfolio_value = initial_portfolio_value\n", "\n", "# Set the transaction cost (e.g., commission, slippage)\n", "transaction_cost = 0.001\n", "\n", "# Perform backtesting on the test data\n", "for state in range(len(test_states)):\n", " # Choose the action with the highest Q-value\n", " action = actions[np.argmax(q_table_train[state])]\n", " \n", " # Take the action and observe the price change and reward\n", " stock_index = np.random.randint(0, test_states.shape[1])\n", " price_change = test_states[state, stock_index]\n", " reward = calculate_reward(action, price_change)\n", " \n", " # Calculate the new portfolio value based on the action and price change\n", " if action == 'buy':\n", " portfolio_value *= (1 + price_change - transaction_cost)\n", " elif action == 'sell':\n", " portfolio_value *= (1 - price_change - transaction_cost)\n", " \n", " # Store the backtesting results\n", " new_result = pd.DataFrame({\n", " 'Action': [action],\n", " 'Price_Change': [price_change],\n", " 'Reward': [reward],\n", " 'Portfolio_Value': [portfolio_value]\n", " })\n", " backtest_results = pd.concat([backtest_results, new_result], ignore_index=True)\n", "\n", "# Calculate the total return\n", "total_return = (portfolio_value - initial_portfolio_value) / initial_portfolio_value\n", "\n", "# Calculate the Sharpe ratio\n", "daily_returns = backtest_results['Portfolio_Value'].pct_change()\n", "sharpe_ratio = np.sqrt(252) * daily_returns.mean() / daily_returns.std()\n", "\n", "# Print the backtesting results and performance metrics\n", "print(\"Backtesting Results:\")\n", "print(backtest_results)\n", "print(\"\\nPerformance Metrics:\")\n", "print(\"Total Return:\", total_return)\n", "print(\"Sharpe Ratio:\", sharpe_ratio)\n", "\n", "# Visualize the backtesting results\n", "plt.figure(figsize=(10, 6))\n", "plt.plot(backtest_results['Portfolio_Value'])\n", "plt.title('Portfolio Value Over Time')\n", "plt.xlabel('Time Step')\n", "plt.ylabel('Portfolio Value')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Part 3.2\n", "Compare the performance of the reinforcement learning-based strategy with baseline strategies (e.g., buy-and-hold)." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Reinforcement Learning Strategy:\n", "Total Return: 0.11098795325480733\n", "Sharpe Ratio: 1.722366460147563\n", "\n", "Buy-and-Hold Strategy:\n", "Total Return: 0.0799976737496936\n", "Sharpe Ratio: 1.7053007904465927\n" ] } ], "source": [ "# Create DataFrames to store the backtesting results\n", "rl_backtest_results = pd.DataFrame(columns=['Action', 'Price_Change', 'Reward', 'Portfolio_Value'])\n", "bh_backtest_results = pd.DataFrame(columns=['Action', 'Price_Change', 'Portfolio_Value'])\n", "\n", "# Set the initial portfolio values\n", "rl_portfolio_value = initial_portfolio_value\n", "bh_portfolio_value = initial_portfolio_value\n", "\n", "# Set the transaction cost (e.g., commission, slippage)\n", "transaction_cost = 0.001\n", "\n", "# Perform backtesting for the reinforcement learning strategy\n", "for state in range(len(test_states)):\n", " # Choose the action with the highest Q-value\n", " action = actions[np.argmax(q_table_train[state])]\n", " \n", " # Take the action and observe the price change and reward\n", " stock_index = np.random.randint(0, test_states.shape[1])\n", " price_change = test_states[state, stock_index]\n", " reward = calculate_reward(action, price_change)\n", " \n", " # Calculate the new portfolio value based on the action and price change\n", " if action == 'buy':\n", " rl_portfolio_value *= (1 + price_change - transaction_cost)\n", " elif action == 'sell':\n", " rl_portfolio_value *= (1 - price_change - transaction_cost)\n", " \n", " # Store the backtesting results for the reinforcement learning strategy\n", " new_rl_result = pd.DataFrame({\n", " 'Action': [action],\n", " 'Price_Change': [price_change],\n", " 'Reward': [reward],\n", " 'Portfolio_Value': [rl_portfolio_value]\n", " })\n", " rl_backtest_results = pd.concat([rl_backtest_results, new_rl_result], ignore_index=True)\n", "\n", "# Perform backtesting for the buy-and-hold strategy\n", "for state in range(len(test_states)):\n", " # Assume the buy-and-hold strategy always holds the stock\n", " action = 'hold'\n", " \n", " # Observe the price change\n", " stock_index = np.random.randint(0, test_states.shape[1])\n", " price_change = test_states[state, stock_index]\n", " \n", " # Calculate the new portfolio value based on the price change\n", " bh_portfolio_value *= (1 + price_change)\n", " \n", " # Store the backtesting results for the buy-and-hold strategy\n", " new_bh_result = pd.DataFrame({\n", " 'Action': [action],\n", " 'Price_Change': [price_change],\n", " 'Portfolio_Value': [bh_portfolio_value]\n", " })\n", " bh_backtest_results = pd.concat([bh_backtest_results, new_bh_result], ignore_index=True)\n", "\n", "# Calculate the total returns\n", "rl_total_return = (rl_portfolio_value - initial_portfolio_value) / initial_portfolio_value\n", "bh_total_return = (bh_portfolio_value - initial_portfolio_value) / initial_portfolio_value\n", "\n", "# Calculate the Sharpe ratios\n", "rl_daily_returns = rl_backtest_results['Portfolio_Value'].pct_change()\n", "rl_sharpe_ratio = np.sqrt(252) * rl_daily_returns.mean() / rl_daily_returns.std()\n", "\n", "bh_daily_returns = bh_backtest_results['Portfolio_Value'].pct_change()\n", "bh_sharpe_ratio = np.sqrt(252) * bh_daily_returns.mean() / bh_daily_returns.std()\n", "\n", "# Print the performance metrics for both strategies\n", "print(\"Reinforcement Learning Strategy:\")\n", "print(\"Total Return:\", rl_total_return)\n", "print(\"Sharpe Ratio:\", rl_sharpe_ratio)\n", "\n", "print(\"\\nBuy-and-Hold Strategy:\")\n", "print(\"Total Return:\", bh_total_return)\n", "print(\"Sharpe Ratio:\", bh_sharpe_ratio)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "By comparing the total returns and Sharpe ratios, we can assess the relative performance of the reinforcement learning strategy against the buy-and-hold baseline. If the reinforcement learning strategy has a higher total return and Sharpe ratio, it suggests that it outperforms the buy-and-hold strategy on both absolute and risk-adjusted bases.\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Part 4.1\n", "Visualize the trading decisions made by the agent over time. Analyze the cumulative returns, Sharpe ratio, maximum drawdown, and other relevant metrics." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Reinforcement Learning Strategy:\n", "Cumulative Returns: 0.10675032975910395\n", "Maximum Drawdown: -0.08461035028173602\n", "\n", "Buy-and-Hold Strategy:\n", "Cumulative Returns: 0.09097306322775656\n", "Maximum Drawdown: -0.09308351791831726\n" ] } ], "source": [ "# Visualize the trading decisions and portfolio values over time\n", "plt.figure(figsize=(12, 8))\n", "plt.subplot(2, 1, 1)\n", "plt.plot(rl_backtest_results.index, rl_backtest_results['Portfolio_Value'], label='Reinforcement Learning')\n", "plt.plot(bh_backtest_results.index, bh_backtest_results['Portfolio_Value'], label='Buy-and-Hold')\n", "plt.title('Portfolio Value Over Time')\n", "plt.xlabel('Time')\n", "plt.ylabel('Portfolio Value')\n", "plt.legend()\n", "\n", "plt.subplot(2, 1, 2)\n", "plt.plot(rl_backtest_results.index, rl_backtest_results['Action'].map({'buy': 1, 'hold': 0, 'sell': -1}), marker='o', linestyle='None', label='Reinforcement Learning')\n", "plt.title('Trading Decisions Over Time')\n", "plt.xlabel('Time')\n", "plt.ylabel('Action')\n", "plt.legend()\n", "\n", "plt.tight_layout()\n", "plt.show()\n", "\n", "# Calculate additional performance metrics\n", "rl_cumulative_returns = (rl_backtest_results['Portfolio_Value'].iloc[-1] - rl_backtest_results['Portfolio_Value'].iloc[0]) / rl_backtest_results['Portfolio_Value'].iloc[0]\n", "bh_cumulative_returns = (bh_backtest_results['Portfolio_Value'].iloc[-1] - bh_backtest_results['Portfolio_Value'].iloc[0]) / bh_backtest_results['Portfolio_Value'].iloc[0]\n", "\n", "rl_max_drawdown = (rl_backtest_results['Portfolio_Value'] / rl_backtest_results['Portfolio_Value'].cummax() - 1).min()\n", "bh_max_drawdown = (bh_backtest_results['Portfolio_Value'] / bh_backtest_results['Portfolio_Value'].cummax() - 1).min()\n", "\n", "# Print additional performance metrics\n", "print(\"Reinforcement Learning Strategy:\")\n", "print(\"Cumulative Returns:\", rl_cumulative_returns)\n", "print(\"Maximum Drawdown:\", rl_max_drawdown)\n", "\n", "print(\"\\nBuy-and-Hold Strategy:\")\n", "print(\"Cumulative Returns:\", bh_cumulative_returns)\n", "print(\"Maximum Drawdown:\", bh_max_drawdown)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "By visualizing the trading decisions and portfolio values over time, we can gain insights into how the reinforcement learning strategy behaves compared to the buy-and-hold strategy. The plots will show the timing and impact of the trading decisions on the portfolio value." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Part 4.2\n", "Provide insights into the strengths and weaknesses of the developed trading strategy\n", "\n", "\n", "Strengths of the Reinforcement Learning Strategy:\n", "- Higher Cumulative Returns: The reinforcement learning strategy has generated cumulative returns of that outperformed the buy-and-hold strategy, indicating that the reinforcement learning strategy has been more profitable overall during the backtesting period.\n", "- Adaptability: Reinforcement learning algorithms have the ability to learn and adapt to changing market conditions. T\n", "\n", "Weaknesses of the Reinforcement Learning Strategy:\n", "- Maximum Drawdown: While the reinforcement learning strategy has a lower drawdown, it still indicates potential risks and the need for effective risk management techniques.\n", "- Complexity and Computational Requirements: Reinforcement learning algorithms can be complex and computationally intensive.\n", "\n", "\n", "Strengths of the Buy-and-Hold Strategy:\n", "- Simplicity: The buy-and-hold strategy remains straightforward and easy to implement. It does not require active trading decisions or complex algorithms.\n", "- Lower Maximum Drawdown: The buy-and-hold strategy has experienced a slightly higher maximum drawdown, however, the difference is relatively small, and both strategies have faced significant drawdowns.\n", "\n", "Weaknesses of the Buy-and-Hold Strategy:\n", "- Lower Cumulative Returns: The buy-and-hold strategy has generated lower cumulative returns suggesting that the passive approach of holding the asset has been less profitable during the backtesting period.\n", "- Lack of Adaptability: The buy-and-hold strategy does not actively adapt to changing market conditions. It relies on the long-term growth of the asset and may miss out on short-term trading opportunities." ] }, { "cell_type": "markdown", "metadata": {}, "source": [] } ], "metadata": { "kernelspec": { "display_name": "dev2", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.10.11" } }, "nbformat": 4, "nbformat_minor": 2 }