sma-agent-backtesting.ipynb

sma-agent-backtesting.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyO5xx25x39Ls8ItrGZh6PYz",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/virattt/b624560b05f50b69f1042a7a4f4adceb/sma-agent-backtesting.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "jDTXcDmUj0bV"
      },
      "outputs": [],
      "source": [
        "!pip install -U --quiet langgraph langchain_openai"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import getpass\n",
        "import os\n",
        "\n",
        "\n",
        "def _set_if_undefined(var: str):\n",
        "    if not os.environ.get(var):\n",
        "        os.environ[var] = getpass.getpass(f\"Please provide your {var}\")\n",
        "\n",
        "\n",
        "_set_if_undefined(\"FINANCIAL_DATASETS_API_KEY\")   # For getting financial data. Get from https://financialdatasets.ai\n",
        "_set_if_undefined(\"OPENAI_API_KEY\")   # For getting financial data. Get from https://financialdatasets.ai"
      ],
      "metadata": {
        "id": "azDJXqYSl2sF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import requests\n",
        "import os\n",
        "from datetime import timedelta\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Import your agent's dependencies\n",
        "from langchain_openai.chat_models import ChatOpenAI\n",
        "from langchain_core.messages import HumanMessage, SystemMessage\n",
        "from langgraph.graph import StateGraph, MessagesState"
      ],
      "metadata": {
        "id": "lP3IylZfm744"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 1. Create the Agent"
      ],
      "metadata": {
        "id": "-adeYrdHpHvx"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Initialize the OpenAI model\n",
        "gpt_4o_model = ChatOpenAI(model=\"gpt-4o\", temperature=0)\n",
        "\n",
        "# Update the system prompt to include portfolio management\n",
        "system_prompt = \"\"\"\n",
        "You are a financial trading agent using a Simple Moving Average (SMA) crossover strategy with portfolio management capabilities.\n",
        "\n",
        "Your responsibilities:\n",
        "1. Analyze technical indicators (5-day and 20-day SMAs)\n",
        "2. Manage portfolio allocation based on:\n",
        "   - Available cash\n",
        "   - Current stock position\n",
        "   - Current market price\n",
        "3. Generate specific trading orders with:\n",
        "   - Action: 'buy', 'sell', or 'hold'\n",
        "   - Quantity: number of shares to trade (must be affordable for buys or available for sells)\n",
        "\n",
        "Base your decisions on:\n",
        "- SMA crossover signals (5-day vs 20-day)\n",
        "- Current portfolio state\n",
        "- Risk management (don't use more than 50% of available cash in a single trade)\n",
        "- Trading constraints:\n",
        "  * For buy orders: ensure quantity * current_price <= available cash\n",
        "  * For sell orders: ensure quantity <= current stock position\n",
        "\n",
        "Your response should be in the format:\n",
        "{\n",
        "    \"action\": \"<buy|sell|hold>\",\n",
        "    \"quantity\": <number_of_shares>\n",
        "}\n",
        "Only output this JSON object, without any additional text.\n",
        "\"\"\"\n",
        "\n",
        "\n",
        "# Update the function that calls the model\n",
        "def call_agent(state: MessagesState):\n",
        "    prompt = SystemMessage(content=system_prompt)\n",
        "    messages = state[\"messages\"]\n",
        "\n",
        "    if messages and messages[0].content != system_prompt:\n",
        "        messages.insert(0, prompt)\n",
        "\n",
        "    if hasattr(messages[-1], \"additional_kwargs\"):\n",
        "        params = messages[-1].additional_kwargs\n",
        "        historical_data = get_price_data(\n",
        "            params[\"ticker\"], params[\"start_date\"], params[\"end_date\"]\n",
        "        )\n",
        "        signals = calculate_trading_signals(historical_data)\n",
        "        portfolio = params.get(\"portfolio\", {\n",
        "            \"cash\": params.get(\"initial_capital\", 100000),\n",
        "            \"stock\": 0\n",
        "        })\n",
        "\n",
        "        messages[-1].content = f\"\"\"\n",
        "        Current price: {signals['current_price']:.2f}\n",
        "        5-day SMA: {signals['sma_5_curr']:.2f} (previous: {signals['sma_5_prev']:.2f})\n",
        "        20-day SMA: {signals['sma_20_curr']:.2f} (previous: {signals['sma_20_prev']:.2f})\n",
        "\n",
        "        Portfolio:\n",
        "        Cash: ${portfolio['cash']:.2f}\n",
        "        Shares: {portfolio['stock']}\n",
        "\n",
        "        Based on the SMA crossover strategy and current portfolio, what is your trading decision?\n",
        "        \"\"\"\n",
        "\n",
        "    return {\"messages\": [gpt_4o_model.invoke(messages)]}\n",
        "\n",
        "\n",
        "# Define the agent graph\n",
        "workflow = StateGraph(MessagesState)\n",
        "workflow.add_node(\"agent\", call_agent)\n",
        "workflow.set_entry_point(\"agent\")\n",
        "app = workflow.compile()\n",
        "\n",
        "\n",
        "# Update the run_agent function to include portfolio state\n",
        "def run_agent(ticker: str, start_date: str, end_date: str, portfolio: dict):\n",
        "    final_state = app.invoke(\n",
        "        {\n",
        "            \"messages\": [\n",
        "                HumanMessage(\n",
        "                    content=\"Make a trading decision based on the provided data.\",\n",
        "                    additional_kwargs={\n",
        "                        \"ticker\": ticker,\n",
        "                        \"start_date\": start_date,\n",
        "                        \"end_date\": end_date,\n",
        "                        \"portfolio\": portfolio\n",
        "                    },\n",
        "                )\n",
        "            ]\n",
        "        },\n",
        "        config={\"configurable\": {\"thread_id\": 42}},\n",
        "    )\n",
        "    return final_state[\"messages\"][-1].content"
      ],
      "metadata": {
        "id": "tuMK7Jrxj3eY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "run_agent(\"hello\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "UnlmOdL8j7QD",
        "outputId": "4197be79-43fb-440f-f1dd-f2fdd5d9b282"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'hold'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 2. Get Price Data"
      ],
      "metadata": {
        "id": "DsvZo7gjpMGu"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def get_price_data(ticker, start_date, end_date):\n",
        "    # Add your API key to the headers\n",
        "    headers = {\n",
        "        \"X-API-KEY\": os.environ.get(\"FINANCIAL_DATASETS_API_KEY\")\n",
        "    }\n",
        "\n",
        "    # Create the URL\n",
        "    url = (\n",
        "        f'https://api.financialdatasets.ai/prices/'\n",
        "        f'?ticker={ticker}'\n",
        "        f'&interval=day'\n",
        "        f'&interval_multiplier=1'\n",
        "        f'&start_date={start_date}'\n",
        "        f'&end_date={end_date}'\n",
        "    )\n",
        "\n",
        "    # Make API request\n",
        "    response = requests.get(url, headers=headers)\n",
        "\n",
        "    # Check for successful response\n",
        "    if response.status_code != 200:\n",
        "        raise Exception(f\"Error fetching data: {response.status_code} - {response.text}\")\n",
        "\n",
        "    # Parse prices from the response\n",
        "    data = response.json()\n",
        "    prices = data.get('prices')\n",
        "    if not prices:\n",
        "        raise ValueError(\"No price data returned\")\n",
        "\n",
        "    # Convert prices to DataFrame\n",
        "    df = pd.DataFrame(prices)\n",
        "\n",
        "    # Convert 'time' to datetime and set as index\n",
        "    df['Date'] = pd.to_datetime(df['time'])\n",
        "    df.set_index('Date', inplace=True)\n",
        "\n",
        "    # Ensure numeric data types\n",
        "    numeric_cols = ['open', 'close', 'high', 'low', 'volume']\n",
        "    for col in numeric_cols:\n",
        "        df[col] = pd.to_numeric(df[col], errors='coerce')\n",
        "\n",
        "    # Sort by date\n",
        "    df.sort_index(inplace=True)\n",
        "\n",
        "    return df\n"
      ],
      "metadata": {
        "id": "tCYkY3EDmCLM"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 3. Define Trading Strategy"
      ],
      "metadata": {
        "id": "4R8X9AgfKhoz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Define a function to calculate trading signals\n",
        "def calculate_trading_signals(historical_data: pd.DataFrame) -> dict:\n",
        "    \"\"\"Calculate trading signals based on SMA crossover strategy\"\"\"\n",
        "    # Calculate SMAs\n",
        "    sma_5 = historical_data[\"close\"].rolling(window=5).mean()\n",
        "    sma_20 = historical_data[\"close\"].rolling(window=20).mean()\n",
        "\n",
        "    # Get the last two points of each SMA to check for crossover\n",
        "    sma_5_prev, sma_5_curr = sma_5.iloc[-2:]\n",
        "    sma_20_prev, sma_20_curr = sma_20.iloc[-2:]\n",
        "\n",
        "    return {\n",
        "        \"current_price\": historical_data[\"close\"].iloc[-1],\n",
        "        \"sma_5_curr\": sma_5_curr,\n",
        "        \"sma_5_prev\": sma_5_prev,\n",
        "        \"sma_20_curr\": sma_20_curr,\n",
        "        \"sma_20_prev\": sma_20_prev,\n",
        "    }"
      ],
      "metadata": {
        "id": "25BbhA_FKj3k"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 4. Create a backtester"
      ],
      "metadata": {
        "id": "eib4alsrpQ3h"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class Backtester:\n",
        "    def __init__(self, agent, ticker, start_date, end_date, initial_capital):\n",
        "        self.agent = agent\n",
        "        self.ticker = ticker\n",
        "        self.start_date = start_date\n",
        "        self.end_date = end_date\n",
        "        self.initial_capital = initial_capital\n",
        "        self.portfolio = {\"cash\": initial_capital, \"stock\": 0}\n",
        "        self.portfolio_values = []\n",
        "\n",
        "    def parse_action(self, agent_output):\n",
        "        try:\n",
        "            # Expect JSON output from agent\n",
        "            import json\n",
        "            decision = json.loads(agent_output)\n",
        "            return decision[\"action\"], decision[\"quantity\"]\n",
        "        except:\n",
        "            return \"hold\", 0\n",
        "\n",
        "    def execute_trade(self, action, quantity, current_price):\n",
        "        \"\"\"Validate and execute trades based on portfolio constraints\"\"\"\n",
        "        if action == \"buy\" and quantity > 0:\n",
        "            cost = quantity * current_price\n",
        "            if cost <= self.portfolio[\"cash\"]:\n",
        "                self.portfolio[\"stock\"] += quantity\n",
        "                self.portfolio[\"cash\"] -= cost\n",
        "                return quantity\n",
        "            else:\n",
        "                # Calculate maximum affordable quantity\n",
        "                max_quantity = self.portfolio[\"cash\"] // current_price\n",
        "                if max_quantity > 0:\n",
        "                    self.portfolio[\"stock\"] += max_quantity\n",
        "                    self.portfolio[\"cash\"] -= max_quantity * current_price\n",
        "                    return max_quantity\n",
        "                return 0\n",
        "        elif action == \"sell\" and quantity > 0:\n",
        "            quantity = min(quantity, self.portfolio[\"stock\"])\n",
        "            if quantity > 0:\n",
        "                self.portfolio[\"cash\"] += quantity * current_price\n",
        "                self.portfolio[\"stock\"] -= quantity\n",
        "                return quantity\n",
        "            return 0\n",
        "        return 0\n",
        "\n",
        "    def run_backtest(self):\n",
        "        dates = pd.date_range(self.start_date, self.end_date, freq=\"B\")\n",
        "\n",
        "        print(\"\\nStarting backtest...\")\n",
        "        print(f\"{'Date':<12} {'Action':<6} {'Quantity':>8} {'Price':>8} {'Cash':>12} {'Stock':>8} {'Total Value':>12}\")\n",
        "        print(\"-\" * 70)\n",
        "\n",
        "        for current_date in dates:\n",
        "            lookback_start = (current_date - timedelta(days=30)).strftime(\"%Y-%m-%d\")\n",
        "            current_date_str = current_date.strftime(\"%Y-%m-%d\")\n",
        "\n",
        "            agent_output = self.agent(\n",
        "                ticker=self.ticker,\n",
        "                start_date=lookback_start,\n",
        "                end_date=current_date_str,\n",
        "                portfolio=self.portfolio\n",
        "            )\n",
        "\n",
        "            action, quantity = self.parse_action(agent_output)\n",
        "            df = get_price_data(self.ticker, lookback_start, current_date_str)\n",
        "            current_price = df.iloc[-1]['close']\n",
        "\n",
        "            # Execute the trade with validation\n",
        "            executed_quantity = self.execute_trade(action, quantity, current_price)\n",
        "\n",
        "            # Update total portfolio value\n",
        "            total_value = self.portfolio[\"cash\"] + self.portfolio[\"stock\"] * current_price\n",
        "            self.portfolio[\"portfolio_value\"] = total_value\n",
        "\n",
        "            # Log the current state with executed quantity\n",
        "            print(\n",
        "                f\"{current_date.strftime('%Y-%m-%d'):<12} {action:<6} {executed_quantity:>8} {current_price:>8.2f} \"\n",
        "                f\"{self.portfolio['cash']:>12.2f} {self.portfolio['stock']:>8} {total_value:>12.2f}\"\n",
        "            )\n",
        "\n",
        "            # Record the portfolio value\n",
        "            self.portfolio_values.append(\n",
        "                {\"Date\": current_date, \"Portfolio Value\": total_value}\n",
        "            )\n",
        "\n",
        "    def analyze_performance(self):\n",
        "        # Convert portfolio values to DataFrame\n",
        "        performance_df = pd.DataFrame(self.portfolio_values).set_index(\"Date\")\n",
        "\n",
        "        # Calculate total return\n",
        "        total_return = (\n",
        "            self.portfolio[\"portfolio_value\"] - self.initial_capital\n",
        "        ) / self.initial_capital\n",
        "        print(f\"Total Return: {total_return * 100:.2f}%\")\n",
        "\n",
        "        # Plot the portfolio value over time\n",
        "        performance_df[\"Portfolio Value\"].plot(\n",
        "            title=\"Portfolio Value Over Time\", figsize=(12, 6)\n",
        "        )\n",
        "        plt.ylabel(\"Portfolio Value ($)\")\n",
        "        plt.xlabel(\"Date\")\n",
        "        plt.show()\n",
        "\n",
        "        # Compute daily returns\n",
        "        performance_df[\"Daily Return\"] = performance_df[\"Portfolio Value\"].pct_change()\n",
        "\n",
        "        # Calculate Sharpe Ratio (assuming 252 trading days in a year)\n",
        "        mean_daily_return = performance_df[\"Daily Return\"].mean()\n",
        "        std_daily_return = performance_df[\"Daily Return\"].std()\n",
        "        sharpe_ratio = (mean_daily_return / std_daily_return) * (252**0.5)\n",
        "        print(f\"Sharpe Ratio: {sharpe_ratio:.2f}\")\n",
        "\n",
        "        # Calculate Maximum Drawdown\n",
        "        rolling_max = performance_df[\"Portfolio Value\"].cummax()\n",
        "        drawdown = performance_df[\"Portfolio Value\"] / rolling_max - 1\n",
        "        max_drawdown = drawdown.min()\n",
        "        print(f\"Maximum Drawdown: {max_drawdown * 100:.2f}%\")\n",
        "\n",
        "        return performance_df"
      ],
      "metadata": {
        "id": "sycJCgYunBfq"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 4. Run the Backtest"
      ],
      "metadata": {
        "id": "8gbjCJT-pTa0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Define parameters\n",
        "ticker = \"AAPL\"  # Example ticker symbol\n",
        "start_date = \"2024-01-01\"  # Adjust as needed\n",
        "end_date = \"2024-10-31\"  # Adjust as needed\n",
        "initial_capital = 100000  # $100,000\n",
        "\n",
        "# Create an instance of Backtester\n",
        "backtester = Backtester(\n",
        "    agent=run_agent,\n",
        "    ticker=ticker,\n",
        "    start_date=start_date,\n",
        "    end_date=end_date,\n",
        "    initial_capital=initial_capital,\n",
        ")\n",
        "\n",
        "# Run the backtesting process\n",
        "backtester.run_backtest()\n",
        "performance_df = backtester.analyze_performance()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 959
        },
        "id": "ro6_juA_nHl-",
        "outputId": "02f848e0-baa6-4f5d-b0d7-30b27292408e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "Starting backtest...\n",
            "Date         Action    Price   Shares         Cash  Total Value\n",
            "------------------------------------------------------------\n",
            "2024-01-09   buy      185.14      540        24.40    100000.00\n",
            "2024-01-10   hold     186.19      540        24.40    100567.00\n",
            "2024-01-11   buy      185.59      540        24.40    100243.00\n",
            "2024-01-12   hold     185.92      540        24.40    100421.20\n",
            "2024-01-16   hold     183.63      540        24.40     99184.60\n",
            "2024-01-17   sell     182.68        0     98671.60     98671.60\n",
            "2024-01-18   buy      188.63      523        18.11     98671.60\n",
            "2024-01-19   buy      191.56      523        18.11    100203.99\n",
            "2024-01-22   buy      193.89      523        18.11    101422.58\n",
            "2024-01-23   buy      195.18      523        18.11    102097.25\n",
            "2024-01-24   hold     194.50      523        18.11    101741.61\n",
            "2024-01-25   hold     194.17      523        18.11    101569.02\n",
            "2024-01-26   hold     192.42      523        18.11    100653.77\n",
            "2024-01-29   sell     191.73        0    100292.90    100292.90\n",
            "2024-01-30   buy      188.04      533        67.58    100292.90\n",
            "2024-01-31   buy      184.40      533        67.58     98352.78\n",
            "Total Return: -1.65%\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Sharpe Ratio: -1.89\n",
            "Maximum Drawdown: -3.67%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "3qmllXP2kMH4"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}