From 4bc947cf07806b68919fd6a3ab86352ff063a560 Mon Sep 17 00:00:00 2001 From: Oleg Sheynin Date: Tue, 15 Jul 2025 19:24:18 +0000 Subject: [PATCH] progress --- research/notebooks/pt_sliding.ipynb | 2210 ++++++++++++++------------- 1 file changed, 1123 insertions(+), 1087 deletions(-) diff --git a/research/notebooks/pt_sliding.ipynb b/research/notebooks/pt_sliding.ipynb index e0d6636..60508bf 100644 --- a/research/notebooks/pt_sliding.ipynb +++ b/research/notebooks/pt_sliding.ipynb @@ -43,7 +43,7 @@ }, { "cell_type": "code", - "execution_count": 47, + "execution_count": 13, "metadata": {}, "outputs": [], "source": [ @@ -51,6 +51,17 @@ "# Specify your configuration file, trading symbols and date here\n", "\n", "# Configuration file selection\n", + "global CONFIG_FILE\n", + "global SYMBOL_A\n", + "global SYMBOL_B\n", + "global TRADING_DATE\n", + "global TRD_DATE\n", + "global PT_BT_CONFIG\n", + "global DATA_FILE\n", + "global FIT_METHOD_TYPE\n", + "\n", + "FIT_METHOD_TYPE = \"SlidingFit\"\n", + "\n", "CONFIG_FILE = \"equity\" # Options: \"equity\", \"crypto\", or custom filename (without .cfg extension)\n", "\n", "# Trading pair symbols\n", @@ -78,44 +89,30 @@ }, { "cell_type": "code", - "execution_count": 48, + "execution_count": 14, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Setup complete!\n" - ] - } - ], + "outputs": [], "source": [ - "import sys\n", - "import os\n", - "sys.path.append('/home/oleg/develop/pairs_trading/lib')\n", - "sys.path.append('/home/coder/pairs_trading/lib')\n", + "def setup() -> None:\n", + " import sys\n", + " import os\n", + " sys.path.append('/home/oleg/develop/pairs_trading/lib')\n", + " sys.path.append('/home/coder/pairs_trading/lib')\n", "\n", - "import pandas as pd\n", - "import numpy as np\n", - "import matplotlib.pyplot as plt\n", - "import seaborn as sns\n", - "import importlib\n", - "from typing import Dict, List, Optional\n", - "from IPython.display import clear_output\n", + " import pandas as pd\n", + " import numpy as np\n", + " import importlib\n", + " from typing import Dict, List, Optional\n", + " from IPython.display import clear_output\n", "\n", - "# Import our modules\n", - "from pt_trading.sliding_fit import SlidingFit\n", - "from pt_trading.fit_method import PairState\n", - "from tools.data_loader import load_market_data\n", - "from pt_trading.trading_pair import TradingPair\n", - "from pt_trading.results import BacktestResult\n", + " # Import our modules\n", + " from pt_trading.sliding_fit import SlidingFit\n", + " from pt_trading.fit_method import PairState\n", + " from pt_trading.trading_pair import TradingPair\n", + " # from pt_trading.results import BacktestResult\n", "\n", - "# Set plotting style\n", - "plt.style.use('seaborn-v0_8')\n", - "sns.set_palette(\"husl\")\n", - "plt.rcParams['figure.figsize'] = (15, 10)\n", "\n", - "print(\"Setup complete!\")\n" + " print(\"Setup complete!\")\n" ] }, { @@ -131,17 +128,20 @@ }, { "cell_type": "code", - "execution_count": 49, + "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "# Load Configuration from Configuration Files using HJSON\n", + "from typing import Dict, Optional\n", "import hjson\n", "import os\n", + "import importlib\n", "\n", - "def load_config_from_file(config_type) -> Optional[Dict]:\n", + "\n", + "def load_config_from_file() -> Optional[Dict]:\n", " \"\"\"Load configuration from configuration files using HJSON\"\"\"\n", - " config_file = f\"../../configuration/{config_type}.cfg\"\n", + " config_file = f\"../../configuration/{CONFIG_FILE}.cfg\"\n", " \n", " try:\n", " with open(config_file, 'r') as f:\n", @@ -208,13 +208,1077 @@ }, { "cell_type": "code", - "execution_count": 50, + "execution_count": 16, + "metadata": {}, + "outputs": [], + "source": [ + "def print_config() -> None:\n", + " global PT_BT_CONFIG\n", + " global CONFIG_FILE\n", + " global SYMBOL_A\n", + " global SYMBOL_B\n", + " global TRD_DATE\n", + " global DATA_FILE\n", + " global FIT_MODEL\n", + "\n", + " print(f\"Trading Parameters:\")\n", + " print(f\" Configuration: {CONFIG_FILE}\")\n", + " print(f\" Symbol A: {SYMBOL_A}\")\n", + " print(f\" Symbol B: {SYMBOL_B}\")\n", + " print(f\" Trading Date: {TRD_DATE}\")\n", + "\n", + " # Load the specified configuration\n", + " print(f\"\\nLoading {CONFIG_FILE} configuration using HJSON...\")\n", + "\n", + " CONFIG = load_config_from_file()\n", + " assert CONFIG is not None\n", + " PT_BT_CONFIG = dict(CONFIG)\n", + "\n", + " if PT_BT_CONFIG:\n", + " print(f\"✓ Successfully loaded {PT_BT_CONFIG['security_type']} configuration\")\n", + " print(f\" Data directory: {PT_BT_CONFIG['data_directory']}\")\n", + " print(f\" Database table: {PT_BT_CONFIG['db_table_name']}\")\n", + " print(f\" Exchange: {PT_BT_CONFIG['exchange_id']}\")\n", + " print(f\" Training window: {PT_BT_CONFIG['training_minutes']} minutes\")\n", + " print(f\" Open threshold: {PT_BT_CONFIG['dis-equilibrium_open_trshld']}\")\n", + " print(f\" Close threshold: {PT_BT_CONFIG['dis-equilibrium_close_trshld']}\")\n", + " \n", + " # Instantiate strategy from config\n", + " FIT_MODEL = instantiate_fit_method_from_config(PT_BT_CONFIG)\n", + " print(f\" Fit Method: {type(FIT_MODEL).__name__}\")\n", + " \n", + " # Automatically construct data file name based on date and config type\n", + " DATA_FILE = f\"{TRADING_DATE}.mktdata.ohlcv.db\"\n", + "\n", + " # Update CONFIG with the specific data file and instruments\n", + " PT_BT_CONFIG[\"datafiles\"] = [DATA_FILE]\n", + " PT_BT_CONFIG[\"instruments\"] = [SYMBOL_A, SYMBOL_B]\n", + " \n", + " print(f\"\\nData Configuration:\")\n", + " print(f\" Data File: {DATA_FILE}\")\n", + " print(f\" Security Type: {PT_BT_CONFIG['security_type']}\")\n", + " \n", + " # Verify data file exists\n", + " data_file_path = f\"{PT_BT_CONFIG['data_directory']}/{DATA_FILE}\"\n", + " if os.path.exists(data_file_path):\n", + " print(f\" ✓ Data file found: {data_file_path}\")\n", + " else:\n", + " print(f\" ⚠ Data file not found: {data_file_path}\")\n", + " print(f\" Please check if the date and file exist in the data directory\")\n", + " \n", + " # List available files in the data directory\n", + " try:\n", + " data_dir = PT_BT_CONFIG['data_directory']\n", + " if os.path.exists(data_dir):\n", + " available_files = [f for f in os.listdir(data_dir) if f.endswith('.db')]\n", + " print(f\" Available files in {data_dir}:\")\n", + " for file in sorted(available_files)[:5]: # Show first 5 files\n", + " print(f\" - {file}\")\n", + " if len(available_files) > 5:\n", + " print(f\" ... and {len(available_files)-5} more files\")\n", + " except Exception as e:\n", + " print(f\" Could not list files in data directory: {e}\")\n", + " else:\n", + " print(\"⚠ Failed to load configuration. Please check the configuration file.\")\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "vscode": { + "languageId": "raw" + } + }, + "source": [ + "## Prepare Market Data" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "metadata": {}, + "outputs": [], + "source": [ + "def prepare_market_data() -> None: # Load market data\n", + " global PT_BT_CONFIG\n", + " global DATA_FILE\n", + " global SYMBOL_A\n", + " global SYMBOL_B\n", + " global pair\n", + "\n", + " from tools.data_loader import load_market_data\n", + " from pt_trading.trading_pair import TradingPair\n", + "\n", + "\n", + " datafile_path = f\"{PT_BT_CONFIG['data_directory']}/{DATA_FILE}\"\n", + " print(f\"Loading data from: {datafile_path}\")\n", + "\n", + " market_data_df = load_market_data(datafile_path, config=PT_BT_CONFIG)\n", + "\n", + " print(f\"Loaded {len(market_data_df)} rows of market data\")\n", + " print(f\"Symbols in data: {market_data_df['symbol'].unique()}\")\n", + " print(f\"Time range: {market_data_df['tstamp'].min()} to {market_data_df['tstamp'].max()}\")\n", + "\n", + " # Create trading pair\n", + " pair = TradingPair(\n", + " market_data=market_data_df,\n", + " symbol_a=SYMBOL_A,\n", + " symbol_b=SYMBOL_B,\n", + " price_column=PT_BT_CONFIG[\"price_column\"]\n", + " )\n", + "\n", + " print(f\"\\nCreated trading pair: {pair}\")\n", + " print(f\"Market data shape: {pair.market_data_.shape}\")\n", + " print(f\"Column names: {pair.colnames()}\")\n", + "\n", + " # Display sample data\n", + " print(f\"\\nSample data:\")\n", + " display(pair.market_data_.head())\n", + "\n", + " TRADING_DATE = f\"{pair.market_data_['tstamp'].min()} to {pair.market_data_['tstamp'].max()}\"\n", + "\n", + "# prepare_market_data()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "vscode": { + "languageId": "raw" + } + }, + "source": [ + "## Print Strategy Specifics\n" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": {}, + "outputs": [], + "source": [ + "global FIT_MODEL\n", + "global PT_BT_CONFIG\n", + "global pair\n", + "\n", + "def print_strategy_specifics() -> None: # Determine analysis approach based on strategy type\n", + " print(f\"Analysis for SlidingFit ...\")\n", + "\n", + " print(\"\\n=== SLIDING FIT FIT_MODEL ANALYSIS ===\")\n", + " print(\"This strategy:\")\n", + " print(\" - Re-fits cointegration model using sliding window\")\n", + " print(\" - Adapts to changing market conditions\")\n", + " print(\" - Dynamic parameter updates every minute\")\n", + "\n", + " # Calculate maximum possible iterations for sliding window\n", + " training_minutes = PT_BT_CONFIG[\"training_minutes\"]\n", + " max_iterations = len(pair.market_data_) - training_minutes\n", + " print(f\"\\nSliding window analysis parameters:\")\n", + " print(f\" Training window size: {training_minutes} minutes\")\n", + " print(f\" Maximum iterations: {max_iterations}\")\n", + " print(f\" Total analysis time: ~{max_iterations} minutes\")\n", + "\n", + " print(f\"\\nStrategy Configuration:\")\n", + " print(f\" Open threshold: {PT_BT_CONFIG['dis-equilibrium_open_trshld']}\")\n", + " print(f\" Close threshold: {PT_BT_CONFIG['dis-equilibrium_close_trshld']}\")\n", + " print(f\" Training minutes: {PT_BT_CONFIG['training_minutes']}\")\n", + " print(f\" Funding per pair: ${PT_BT_CONFIG['funding_per_pair']}\")\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "vscode": { + "languageId": "raw" + } + }, + "source": [ + "## Visualize Raw Price Data\n" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": {}, + "outputs": [], + "source": [ + "def visualize_prices() -> None:\n", + " # Plot raw price data\n", + " global price_data\n", + " \n", + " import matplotlib.pyplot as plt\n", + " # Set plotting style\n", + " import seaborn as sns\n", + "\n", + " plt.style.use('seaborn-v0_8')\n", + " sns.set_palette(\"husl\")\n", + " plt.rcParams['figure.figsize'] = (15, 10)\n", + "\n", + " # Get column names for the trading pair\n", + " colname_a, colname_b = pair.colnames()\n", + " price_data = pair.market_data_.copy()\n", + "\n", + " # # 1. Price data - separate plots for each symbol\n", + " # colname_a, colname_b = pair.colnames()\n", + " # price_data = pair.market_data_.copy()\n", + "\n", + " # Create separate subplots for better visibility\n", + " fig_price, price_axes = plt.subplots(2, 1, figsize=(18, 10))\n", + "\n", + " # Plot SYMBOL_A\n", + " price_axes[0].plot(price_data['tstamp'], price_data[colname_a], alpha=0.7, \n", + " label=f'{SYMBOL_A}', linewidth=1, color='blue')\n", + " price_axes[0].set_title(f'{SYMBOL_A} Price Data ({TRD_DATE})')\n", + " price_axes[0].set_ylabel(f'{SYMBOL_A} Price')\n", + " price_axes[0].legend()\n", + " price_axes[0].grid(True)\n", + "\n", + " # Plot SYMBOL_B\n", + " price_axes[1].plot(price_data['tstamp'], price_data[colname_b], alpha=0.7, \n", + " label=f'{SYMBOL_B}', linewidth=1, color='red')\n", + " price_axes[1].set_title(f'{SYMBOL_B} Price Data ({TRD_DATE})')\n", + " price_axes[1].set_ylabel(f'{SYMBOL_B} Price')\n", + " price_axes[1].set_xlabel('Time')\n", + " price_axes[1].legend()\n", + " price_axes[1].grid(True)\n", + "\n", + " plt.tight_layout()\n", + " plt.show()\n", + " \n", + "\n", + " # Plot individual prices\n", + " fig, axes = plt.subplots(2, 1, figsize=(18, 12))\n", + "\n", + " # Normalized prices for comparison\n", + " norm_a = price_data[colname_a] / price_data[colname_a].iloc[0]\n", + " norm_b = price_data[colname_b] / price_data[colname_b].iloc[0]\n", + "\n", + " axes[0].plot(price_data['tstamp'], norm_a, label=f'{SYMBOL_A} (normalized)', alpha=0.8, linewidth=1)\n", + " axes[0].plot(price_data['tstamp'], norm_b, label=f'{SYMBOL_B} (normalized)', alpha=0.8, linewidth=1)\n", + " axes[0].set_title(f'Normalized Price Comparison (Base = 1.0) ({TRD_DATE})')\n", + " axes[0].set_ylabel('Normalized Price')\n", + " axes[0].legend()\n", + " axes[0].grid(True)\n", + "\n", + " # Price ratio\n", + " price_ratio = price_data[colname_a] / price_data[colname_b]\n", + " axes[1].plot(price_data['tstamp'], price_ratio, label=f'{SYMBOL_A}/{SYMBOL_B} Ratio', color='green', alpha=0.8, linewidth=1)\n", + " axes[1].set_title(f'Price Ratio Px({SYMBOL_A})/Px({SYMBOL_B}) ({TRD_DATE})')\n", + " axes[1].set_ylabel('Ratio')\n", + " axes[1].set_xlabel('Time')\n", + " axes[1].legend()\n", + " axes[1].grid(True)\n", + "\n", + " plt.tight_layout()\n", + " plt.show()\n", + "\n", + " # Print basic statistics\n", + " print(f\"\\nPrice Statistics:\")\n", + " print(f\" {SYMBOL_A}: Mean=${price_data[colname_a].mean():.2f}, Std=${price_data[colname_a].std():.2f}\")\n", + " print(f\" {SYMBOL_B}: Mean=${price_data[colname_b].mean():.2f}, Std=${price_data[colname_b].std():.2f}\")\n", + " print(f\" Price Ratio: Mean={price_ratio.mean():.2f}, Std={price_ratio.std():.2f}\")\n", + " print(f\" Correlation: {price_data[colname_a].corr(price_data[colname_b]):.4f}\")\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Analysis" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": {}, + "outputs": [], + "source": [ + " # Initialize strategy state and run analysis\n", + "def run_analysis() -> None:\n", + " global FIT_METHOD_TYPE\n", + " global PT_BT_CONFIG\n", + " global pair\n", + " global FIT_MODEL\n", + " global bt_result\n", + " global pair_trades\n", + "\n", + " import pandas as pd\n", + " from pt_trading.results import BacktestResult\n", + " from pt_trading.fit_method import PairState\n", + "\n", + " print(f\"Running {FIT_METHOD_TYPE} analysis...\")\n", + "\n", + " # Initialize result tracking\n", + " bt_result = BacktestResult(config=PT_BT_CONFIG)\n", + " pair_trades = None\n", + "\n", + " # Run strategy-specific analysis\n", + " print(\"\\n=== SLIDING FIT ANALYSIS ===\")\n", + "\n", + " # Initialize tracking variables for sliding window analysis\n", + " training_minutes = PT_BT_CONFIG[\"training_minutes\"]\n", + " max_iterations = len(pair.market_data_) - training_minutes\n", + "\n", + " # Limit iterations for demonstration (change this for full run)\n", + " max_demo_iterations = min(200, max_iterations)\n", + " print(f\"Processing first {max_demo_iterations} iterations for demonstration...\")\n", + "\n", + " # Initialize pair state for sliding fit method\n", + " pair.user_data_['state'] = PairState.INITIAL\n", + " pair.user_data_[\"trades\"] = pd.DataFrame(columns=pd.Index(FIT_MODEL.TRADES_COLUMNS, dtype=str))\n", + " pair.user_data_[\"is_cointegrated\"] = False\n", + "\n", + " # Run the sliding fit method\n", + " # ==========================================================================\n", + " pair_trades = FIT_MODEL.run_pair(config=PT_BT_CONFIG, pair=pair, bt_result=bt_result)\n", + " # ==========================================================================\n", + "\n", + " if pair_trades is not None and len(pair_trades) > 0:\n", + " print(f\"Generated {len(pair_trades)} trading signals\")\n", + " else:\n", + " print(\"No trading signals generated\")\n", + "\n", + " print(\"\\nStrategy execution completed!\")\n", + "\n", + " # Print comprehensive backtest results\n", + " print(\"\\n\" + \"=\"*80)\n", + " print(\"BACKTEST RESULTS\")\n", + " print(\"=\"*80)\n", + "\n", + " assert pair.predicted_df_ is not None\n", + "\n", + "# run_analysis()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Visualization" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": {}, + "outputs": [], + "source": [ + "def visualization() -> None:\n", + " global price_data\n", + " global pair_trades\n", + " global PT_BT_CONFIG\n", + " global pair\n", + " global SYMBOL_A\n", + " global SYMBOL_B\n", + " global TRD_DATE\n", + "\n", + " import plotly.graph_objects as go\n", + " from plotly.subplots import make_subplots\n", + " import plotly.express as px\n", + " import plotly.offline as pyo\n", + " from IPython.display import HTML\n", + " import pandas as pd\n", + "\n", + " # Configure plotly for offline mode\n", + " pyo.init_notebook_mode(connected=True)\n", + "\n", + " # Strategy-specific interactive visualization\n", + " assert PT_BT_CONFIG is not None\n", + " assert pair.predicted_df_ is not None\n", + "\n", + " print(\"=== SLIDING FIT INTERACTIVE VISUALIZATION ===\")\n", + " print(\"Note: Sliding strategy visualization with interactive plotly charts\")\n", + "\n", + " # Create consistent timeline - superset of timestamps from both dataframes\n", + " market_timestamps = set(pair.market_data_['tstamp'])\n", + " predicted_timestamps = set(pair.predicted_df_['tstamp'])\n", + "\n", + " # Create superset of all timestamps\n", + " all_timestamps = sorted(market_timestamps.union(predicted_timestamps))\n", + "\n", + " # Create a unified timeline dataframe for consistent plotting\n", + " timeline_df = pd.DataFrame({'tstamp': all_timestamps})\n", + "\n", + " # Merge with predicted data to get dis-equilibrium values\n", + " timeline_df = timeline_df.merge(pair.predicted_df_[['tstamp', 'disequilibrium', 'scaled_disequilibrium']], \n", + " on='tstamp', how='left')\n", + "\n", + " # Get Symbol_A and Symbol_B market data\n", + " colname_a, colname_b = pair.colnames()\n", + " symbol_a_data = pair.market_data_[['tstamp', colname_a]].copy()\n", + " symbol_b_data = pair.market_data_[['tstamp', colname_b]].copy()\n", + "\n", + " print(f\"Using consistent timeline with {len(timeline_df)} timestamps\")\n", + " print(f\"Timeline range: {timeline_df['tstamp'].min()} to {timeline_df['tstamp'].max()}\")\n", + "\n", + " # Create subplots with price charts at bottom\n", + " fig = make_subplots(\n", + " rows=4, cols=1,\n", + " subplot_titles=[\n", + " f'Testing Period: Scaled Dis-equilibrium with Trading Thresholds ({TRD_DATE})',\n", + " f'Trading Signal Timeline ({TRD_DATE})',\n", + " f'{SYMBOL_A} Market Data with Trading Signals ({TRD_DATE})',\n", + " f'{SYMBOL_B} Market Data with Trading Signals ({TRD_DATE})'\n", + " ],\n", + " vertical_spacing=0.06,\n", + " specs=[[{\"secondary_y\": False}],\n", + " [{\"secondary_y\": False}],\n", + " [{\"secondary_y\": False}],\n", + " [{\"secondary_y\": False}]]\n", + " )\n", + "\n", + " # 1. Scaled dis-equilibrium with thresholds - using consistent timeline\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=timeline_df['tstamp'],\n", + " y=timeline_df['scaled_disequilibrium'],\n", + " name='Scaled Dis-equilibrium',\n", + " line=dict(color='green', width=2),\n", + " opacity=0.8\n", + " ),\n", + " row=1, col=1\n", + " )\n", + "\n", + " # Add threshold lines to first subplot\n", + " fig.add_shape(\n", + " type=\"line\",\n", + " x0=timeline_df['tstamp'].min(),\n", + " x1=timeline_df['tstamp'].max(),\n", + " y0=PT_BT_CONFIG['dis-equilibrium_open_trshld'],\n", + " y1=PT_BT_CONFIG['dis-equilibrium_open_trshld'],\n", + " line=dict(color=\"purple\", width=2, dash=\"dot\"),\n", + " opacity=0.7,\n", + " row=1, col=1\n", + " )\n", + "\n", + " fig.add_shape(\n", + " type=\"line\",\n", + " x0=timeline_df['tstamp'].min(),\n", + " x1=timeline_df['tstamp'].max(),\n", + " y0=-PT_BT_CONFIG['dis-equilibrium_open_trshld'],\n", + " y1=-PT_BT_CONFIG['dis-equilibrium_open_trshld'],\n", + " line=dict(color=\"purple\", width=2, dash=\"dot\"),\n", + " opacity=0.7,\n", + " row=1, col=1\n", + " )\n", + "\n", + " fig.add_shape(\n", + " type=\"line\",\n", + " x0=timeline_df['tstamp'].min(),\n", + " x1=timeline_df['tstamp'].max(),\n", + " y0=PT_BT_CONFIG['dis-equilibrium_close_trshld'],\n", + " y1=PT_BT_CONFIG['dis-equilibrium_close_trshld'],\n", + " line=dict(color=\"brown\", width=2, dash=\"dot\"),\n", + " opacity=0.7,\n", + " row=1, col=1\n", + " )\n", + "\n", + " fig.add_shape(\n", + " type=\"line\",\n", + " x0=timeline_df['tstamp'].min(),\n", + " x1=timeline_df['tstamp'].max(),\n", + " y0=-PT_BT_CONFIG['dis-equilibrium_close_trshld'],\n", + " y1=-PT_BT_CONFIG['dis-equilibrium_close_trshld'],\n", + " line=dict(color=\"brown\", width=2, dash=\"dot\"),\n", + " opacity=0.7,\n", + " row=1, col=1\n", + " )\n", + "\n", + " fig.add_shape(\n", + " type=\"line\",\n", + " x0=timeline_df['tstamp'].min(),\n", + " x1=timeline_df['tstamp'].max(),\n", + " y0=0,\n", + " y1=0,\n", + " line=dict(color=\"black\", width=1, dash=\"solid\"),\n", + " opacity=0.5,\n", + " row=1, col=1\n", + " )\n", + "\n", + " # 2. Trading signals timeline if available - using consistent timeline\n", + " if pair_trades is not None and len(pair_trades) > 0:\n", + " # Separate trades by action and status for different colors\n", + " buy_open_trades = pair_trades[(pair_trades['action'].str.contains('BUY', na=False)) & \n", + " (pair_trades['status'] == 'OPEN')]\n", + " buy_close_trades = pair_trades[(pair_trades['action'].str.contains('BUY', na=False)) & \n", + " (pair_trades['status'] == 'CLOSE')]\n", + " sell_open_trades = pair_trades[(pair_trades['action'].str.contains('SELL', na=False)) & \n", + " (pair_trades['status'] == 'OPEN')]\n", + " sell_close_trades = pair_trades[(pair_trades['action'].str.contains('SELL', na=False)) & \n", + " (pair_trades['status'] == 'CLOSE')]\n", + " \n", + " # Create y-values for timeline visualization\n", + " trade_indices = list(range(len(pair_trades)))\n", + " \n", + " # Add trading signals with different colors based on action and status\n", + " if len(buy_open_trades) > 0:\n", + " buy_open_indices = [i for i, (_, row) in enumerate(pair_trades.iterrows()) \n", + " if 'BUY' in row['action'] and row['status'] == 'OPEN']\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=buy_open_trades['time'],\n", + " y=buy_open_indices,\n", + " mode='markers',\n", + " name='BUY OPEN',\n", + " marker=dict(color='red', size=10, symbol='circle')\n", + " ),\n", + " row=2, col=1\n", + " )\n", + " \n", + " if len(buy_close_trades) > 0:\n", + " buy_close_indices = [i for i, (_, row) in enumerate(pair_trades.iterrows()) \n", + " if 'BUY' in row['action'] and row['status'] == 'CLOSE']\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=buy_close_trades['time'],\n", + " y=buy_close_indices,\n", + " mode='markers',\n", + " name='BUY CLOSE',\n", + " marker=dict(color='pink', size=10, symbol='circle')\n", + " ),\n", + " row=2, col=1\n", + " )\n", + " \n", + " if len(sell_open_trades) > 0:\n", + " sell_open_indices = [i for i, (_, row) in enumerate(pair_trades.iterrows()) \n", + " if 'SELL' in row['action'] and row['status'] == 'OPEN']\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=sell_open_trades['time'],\n", + " y=sell_open_indices,\n", + " mode='markers',\n", + " name='SELL OPEN',\n", + " marker=dict(color='blue', size=10, symbol='circle')\n", + " ),\n", + " row=2, col=1\n", + " )\n", + " \n", + " if len(sell_close_trades) > 0:\n", + " sell_close_indices = [i for i, (_, row) in enumerate(pair_trades.iterrows()) \n", + " if 'SELL' in row['action'] and row['status'] == 'CLOSE']\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=sell_close_trades['time'],\n", + " y=sell_close_indices,\n", + " mode='markers',\n", + " name='SELL CLOSE',\n", + " marker=dict(color='purple', size=10, symbol='circle')\n", + " ),\n", + " row=2, col=1\n", + " )\n", + "\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=symbol_a_data['tstamp'],\n", + " y=symbol_a_data[colname_a],\n", + " name=f'{SYMBOL_A} Price',\n", + " line=dict(color='blue', width=2),\n", + " opacity=0.8\n", + " ),\n", + " row=3, col=1\n", + " )\n", + "\n", + " if pair_trades is not None and len(pair_trades) > 0:\n", + " # Filter trades for Symbol_A\n", + " symbol_a_trades = pair_trades[pair_trades['symbol'] == SYMBOL_A]\n", + " print(f\"Symbol_A trades: {symbol_a_trades}\")\n", + " \n", + " if len(symbol_a_trades) > 0:\n", + " # Separate trades by action and status for different colors\n", + " buy_open_trades = symbol_a_trades[(symbol_a_trades['action'].str.contains('BUY', na=False)) & \n", + " (symbol_a_trades['status'] == 'OPEN')]\n", + " buy_close_trades = symbol_a_trades[(symbol_a_trades['action'].str.contains('BUY', na=False)) & \n", + " (symbol_a_trades['status'] == 'CLOSE')]\n", + " sell_open_trades = symbol_a_trades[(symbol_a_trades['action'].str.contains('SELL', na=False)) & \n", + " (symbol_a_trades['status'] == 'OPEN')]\n", + " sell_close_trades = symbol_a_trades[(symbol_a_trades['action'].str.contains('SELL', na=False)) & \n", + " (symbol_a_trades['status'] == 'CLOSE')]\n", + " \n", + " # Add BUY OPEN signals\n", + " if len(buy_open_trades) > 0:\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=buy_open_trades['time'],\n", + " y=buy_open_trades['price'],\n", + " mode='markers',\n", + " name=f'{SYMBOL_A} BUY OPEN',\n", + " marker=dict(color='red', size=12, symbol='triangle-up'),\n", + " showlegend=True\n", + " ),\n", + " row=3, col=1\n", + " )\n", + " \n", + " # Add BUY CLOSE signals\n", + " if len(buy_close_trades) > 0:\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=buy_close_trades['time'],\n", + " y=buy_close_trades['price'],\n", + " mode='markers',\n", + " name=f'{SYMBOL_A} BUY CLOSE',\n", + " marker=dict(color='pink', size=12, symbol='triangle-up'),\n", + " showlegend=True\n", + " ),\n", + " row=3, col=1\n", + " )\n", + " \n", + " # Add SELL OPEN signals\n", + " if len(sell_open_trades) > 0:\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=sell_open_trades['time'],\n", + " y=sell_open_trades['price'],\n", + " mode='markers',\n", + " name=f'{SYMBOL_A} SELL OPEN',\n", + " marker=dict(color='blue', size=12, symbol='triangle-down'),\n", + " showlegend=True\n", + " ),\n", + " row=3, col=1\n", + " )\n", + " \n", + " # Add SELL CLOSE signals\n", + " if len(sell_close_trades) > 0:\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=sell_close_trades['time'],\n", + " y=sell_close_trades['price'],\n", + " mode='markers',\n", + " name=f'{SYMBOL_A} SELL CLOSE',\n", + " marker=dict(color='purple', size=12, symbol='triangle-down'),\n", + " showlegend=True\n", + " ),\n", + " row=3, col=1\n", + " )\n", + " \n", + " # 4. Symbol_B Market Data with Trading Signals\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=symbol_b_data['tstamp'],\n", + " y=symbol_b_data[colname_b],\n", + " name=f'{SYMBOL_B} Price',\n", + " line=dict(color='orange', width=2),\n", + " opacity=0.8\n", + " ),\n", + " row=4, col=1\n", + " )\n", + " \n", + " # Add trading signals for Symbol_B if available\n", + " # if pair_trades is not None and len(pair_trades) > 0:\n", + " # Filter trades for Symbol_B\n", + " symbol_b_trades = pair_trades[pair_trades['symbol'] == SYMBOL_B]\n", + " \n", + " if len(symbol_b_trades) > 0:\n", + " # Separate trades by action and status for different colors\n", + " buy_open_trades = symbol_b_trades[(symbol_b_trades['action'].str.contains('BUY', na=False)) & \n", + " (symbol_b_trades['status'] == 'OPEN')]\n", + " buy_close_trades = symbol_b_trades[(symbol_b_trades['action'].str.contains('BUY', na=False)) & \n", + " (symbol_b_trades['status'] == 'CLOSE')]\n", + " sell_open_trades = symbol_b_trades[(symbol_b_trades['action'].str.contains('SELL', na=False)) & \n", + " (symbol_b_trades['status'] == 'OPEN')]\n", + " sell_close_trades = symbol_b_trades[(symbol_b_trades['action'].str.contains('SELL', na=False)) & \n", + " (symbol_b_trades['status'] == 'CLOSE')]\n", + " \n", + " # Add BUY OPEN signals\n", + " if len(buy_open_trades) > 0:\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=buy_open_trades['time'],\n", + " y=buy_open_trades['price'],\n", + " mode='markers',\n", + " name=f'{SYMBOL_B} BUY OPEN',\n", + " marker=dict(color='red', size=12, symbol='triangle-up'),\n", + " showlegend=True\n", + " ),\n", + " row=4, col=1\n", + " )\n", + " \n", + " # Add BUY CLOSE signals\n", + " if len(buy_close_trades) > 0:\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=buy_close_trades['time'],\n", + " y=buy_close_trades['price'],\n", + " mode='markers',\n", + " name=f'{SYMBOL_B} BUY CLOSE',\n", + " marker=dict(color='red', size=12, symbol='triangle-up'),\n", + " showlegend=True\n", + " ),\n", + " row=4, col=1\n", + " )\n", + " \n", + " # Add SELL OPEN signals\n", + " if len(sell_open_trades) > 0:\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=sell_open_trades['time'],\n", + " y=sell_open_trades['price'],\n", + " mode='markers',\n", + " name=f'{SYMBOL_B} SELL OPEN',\n", + " marker=dict(color='blue', size=12, symbol='triangle-down'),\n", + " showlegend=True\n", + " ),\n", + " row=4, col=1\n", + " )\n", + " \n", + " # Add SELL CLOSE signals\n", + " if len(sell_close_trades) > 0:\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=sell_close_trades['time'],\n", + " y=sell_close_trades['price'],\n", + " mode='markers',\n", + " name=f'{SYMBOL_B} SELL CLOSE',\n", + " marker=dict(color='blue', size=12, symbol='triangle-down'),\n", + " showlegend=True\n", + " ),\n", + " row=4, col=1\n", + " )\n", + " \n", + " # Update layout\n", + " fig.update_layout(\n", + " height=1200,\n", + " title_text=f\"Sliding Fit Strategy Analysis - {SYMBOL_A} & {SYMBOL_B} ({TRD_DATE})\",\n", + " showlegend=True,\n", + " template=\"plotly_white\"\n", + " )\n", + " \n", + " # Update y-axis labels\n", + " fig.update_yaxes(title_text=\"Scaled Dis-equilibrium\", row=1, col=1)\n", + " fig.update_yaxes(title_text=\"Signal Index\", row=2, col=1)\n", + " fig.update_yaxes(title_text=f\"{SYMBOL_A} Price ($)\", row=3, col=1)\n", + " fig.update_yaxes(title_text=f\"{SYMBOL_B} Price ($)\", row=4, col=1)\n", + " \n", + " # Update x-axis labels and ensure consistent time range\n", + " time_range = [timeline_df['tstamp'].min(), timeline_df['tstamp'].max()]\n", + " fig.update_xaxes(range=time_range, row=1, col=1)\n", + " fig.update_xaxes(range=time_range, row=2, col=1)\n", + " fig.update_xaxes(range=time_range, row=3, col=1)\n", + " fig.update_xaxes(title_text=\"Time\", range=time_range, row=4, col=1)\n", + " \n", + " # Display using plotly offline mode\n", + " pyo.iplot(fig)\n", + "\n", + " else:\n", + " print(\"No interactive visualization data available - strategy may not have run successfully\")\n", + "\n", + "\n", + "\n", + " # Calculate normalized prices (base = 1.0)\n", + " norm_a = price_data[colname_a] / price_data[colname_a].iloc[0]\n", + " norm_b = price_data[colname_b] / price_data[colname_b].iloc[0]\n", + "\n", + " # Create the main figure\n", + " fig = go.Figure()\n", + "\n", + " # Add normalized price lines\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=price_data['tstamp'],\n", + " y=norm_a,\n", + " name=f'{SYMBOL_A} (Normalized)',\n", + " line=dict(color='blue', width=2),\n", + " opacity=0.8\n", + " )\n", + " )\n", + "\n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=price_data['tstamp'],\n", + " y=norm_b,\n", + " name=f'{SYMBOL_B} (Normalized)',\n", + " line=dict(color='orange', width=2),\n", + " opacity=0.8,\n", + " )\n", + " )\n", + "\n", + " # Add BUY and SELL signals if available\n", + " if pair_trades is not None and len(pair_trades) > 0:\n", + " # Define signal groups to avoid legend repetition\n", + " signal_groups = {}\n", + " \n", + " # Process all trades and group by signal type (ignore OPEN/CLOSE status)\n", + " for _, trade in pair_trades.iterrows():\n", + " symbol = trade['symbol']\n", + " action = trade['action']\n", + " status = trade['status']\n", + " \n", + " # Create signal group key (without status to combine OPEN/CLOSE)\n", + " signal_key = f\"{symbol} {action}\"\n", + " \n", + " # Find normalized price for this trade\n", + " trade_time = trade['time']\n", + " if symbol == SYMBOL_A:\n", + " closest_idx = price_data['tstamp'].searchsorted(trade_time)\n", + " if closest_idx < len(norm_a):\n", + " norm_price = norm_a.iloc[closest_idx]\n", + " else:\n", + " norm_price = norm_a.iloc[-1]\n", + " else: # SYMBOL_B\n", + " closest_idx = price_data['tstamp'].searchsorted(trade_time)\n", + " if closest_idx < len(norm_b):\n", + " norm_price = norm_b.iloc[closest_idx]\n", + " else:\n", + " norm_price = norm_b.iloc[-1]\n", + " \n", + " # Initialize group if not exists\n", + " if signal_key not in signal_groups:\n", + " signal_groups[signal_key] = {\n", + " 'times': [],\n", + " 'prices': [],\n", + " 'actual_prices': [],\n", + " 'symbol': symbol,\n", + " 'action': action,\n", + " 'status': status\n", + " }\n", + " \n", + " # Add to group\n", + " signal_groups[signal_key]['times'].append(trade_time)\n", + " signal_groups[signal_key]['prices'].append(norm_price)\n", + " signal_groups[signal_key]['actual_prices'].append(trade['price'])\n", + " \n", + " # Add each signal group as a single trace\n", + " for signal_key, group_data in signal_groups.items():\n", + " symbol = group_data['symbol']\n", + " action = group_data['action']\n", + " status = group_data['status']\n", + " \n", + " # Determine marker properties (same for all OPEN/CLOSE of same action)\n", + " if 'BUY' in action:\n", + " # marker_color = 'green' if symbol == SYMBOL_A else 'darkgreen'\n", + " marker_color = 'darkgreen'\n", + " marker_symbol = 'triangle-up'\n", + " marker_size = 14\n", + " else: # SELL\n", + " # marker_color = 'orange' if symbol == SYMBOL_A else 'darkred'\n", + " marker_color = 'darkred'\n", + " marker_symbol = 'triangle-down'\n", + " marker_size = 14\n", + " \n", + " # Create hover text for each point in the group\n", + " hover_texts = []\n", + " for i, (time, norm_price, actual_price) in enumerate(zip(group_data['times'], \n", + " group_data['prices'], \n", + " group_data['actual_prices'])):\n", + " # Find the corresponding trade to get the status for hover text\n", + " trade_info = pair_trades[(pair_trades['time'] == time) & \n", + " (pair_trades['symbol'] == symbol) & \n", + " (pair_trades['action'] == action)]\n", + " if len(trade_info) > 0:\n", + " trade_status = trade_info.iloc[0]['status']\n", + " hover_texts.append(f'{signal_key} {trade_status}
' +\n", + " f'Time: {time}
' +\n", + " f'Normalized Price: {norm_price:.4f}
' +\n", + " f'Actual Price: ${actual_price:.2f}')\n", + " else:\n", + " hover_texts.append(f'{signal_key}
' +\n", + " f'Time: {time}
' +\n", + " f'Normalized Price: {norm_price:.4f}
' +\n", + " f'Actual Price: ${actual_price:.2f}')\n", + " \n", + " fig.add_trace(\n", + " go.Scatter(\n", + " x=group_data['times'],\n", + " y=group_data['prices'],\n", + " mode='markers',\n", + " name=signal_key,\n", + " marker=dict(\n", + " color=marker_color,\n", + " size=marker_size,\n", + " symbol=marker_symbol,\n", + " line=dict(width=2, color='black')\n", + " ),\n", + " showlegend=True,\n", + " hovertemplate='%{text}',\n", + " text=hover_texts\n", + " )\n", + " )\n", + "\n", + " # Update layout\n", + " fig.update_layout(\n", + " title=f'Normalized Price Comparison with BUY/SELL Signals - {SYMBOL_A}&{SYMBOL_B} ({TRD_DATE})',\n", + " xaxis_title='Time',\n", + " yaxis_title='Normalized Price (Base = 1.0)',\n", + " height=600,\n", + " showlegend=True,\n", + " template=\"plotly_white\",\n", + " hovermode='x unified'\n", + " )\n", + "\n", + " # Add horizontal line at y=1.0 for reference\n", + " fig.add_hline(y=1.0, line_dash=\"dash\", line_color=\"gray\", opacity=0.5, \n", + " annotation_text=\"Baseline (1.0)\")\n", + "\n", + " # Display the chart\n", + " fig.show()\n", + "\n", + " print(f\"\\nChart shows:\")\n", + " print(f\"- {SYMBOL_A} and {SYMBOL_B} prices normalized to start at 1.0\")\n", + " print(f\"- BUY signals shown as green triangles pointing up\")\n", + " print(f\"- SELL signals shown as orange triangles pointing down\")\n", + " print(f\"- All BUY signals per symbol grouped together, all SELL signals per symbol grouped together\")\n", + " print(f\"- Hover over markers to see individual trade details (OPEN/CLOSE status)\")\n", + "\n", + " if pair_trades is not None and len(pair_trades) > 0:\n", + " print(f\"- Total signals displayed: {len(pair_trades)}\")\n", + " print(f\"- {SYMBOL_A} signals: {len(pair_trades[pair_trades['symbol'] == SYMBOL_A])}\")\n", + " print(f\"- {SYMBOL_B} signals: {len(pair_trades[pair_trades['symbol'] == SYMBOL_B])}\")\n", + " else:\n", + " print(\"- No trading signals to display\")\n", + "\n", + "# visualization()\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "vscode": { + "languageId": "raw" + } + }, + "source": [ + "## Summary and Analysis\n" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": {}, + "outputs": [], + "source": [ + "def summary_and_analysis() -> None:\n", + " print(\"=\" * 80)\n", + " print(\"PAIRS TRADING BACKTEST SUMMARY\")\n", + " print(\"=\" * 80)\n", + "\n", + " print(f\"\\nPair: {SYMBOL_A} & {SYMBOL_B}\")\n", + " print(f\"Fit Method: {FIT_METHOD_TYPE}\")\n", + " print(f\"Configuration: {CONFIG_FILE}\")\n", + " print(f\"Data file: {DATA_FILE}\")\n", + " print(f\"Trading date: {TRD_DATE}\")\n", + "\n", + " print(f\"\\nStrategy Parameters:\")\n", + " print(f\" Training window: {PT_BT_CONFIG['training_minutes']} minutes\")\n", + " print(f\" Open threshold: {PT_BT_CONFIG['dis-equilibrium_open_trshld']}\")\n", + " print(f\" Close threshold: {PT_BT_CONFIG['dis-equilibrium_close_trshld']}\")\n", + " print(f\" Funding per pair: ${PT_BT_CONFIG['funding_per_pair']}\")\n", + "\n", + " # Strategy-specific summary\n", + " print(f\"\\nSliding Window Analysis:\")\n", + " training_minutes = PT_BT_CONFIG['training_minutes']\n", + " max_iterations = len(pair.market_data_) - training_minutes\n", + " print(f\" Total data points: {len(pair.market_data_)}\")\n", + " print(f\" Maximum iterations: {max_iterations}\")\n", + " print(f\" Analysis type: Dynamic sliding window\")\n", + "\n", + " # Trading signals summary\n", + " if pair_trades is not None and len(pair_trades) > 0:\n", + " print(f\"\\nTrading Signals: {len(pair_trades)} generated\")\n", + " unique_times = pair_trades['time'].unique()\n", + " print(f\" Unique trade times: {len(unique_times)}\")\n", + " \n", + " # Group by action type\n", + " buy_signals = pair_trades[pair_trades['action'].str.contains('BUY', na=False)]\n", + " sell_signals = pair_trades[pair_trades['action'].str.contains('SELL', na=False)]\n", + " \n", + " print(f\" BUY signals: {len(buy_signals)}\")\n", + " print(f\" SELL signals: {len(sell_signals)}\")\n", + " \n", + " # Show first few trades\n", + " NTRADES_TO_SHOW = 6\n", + " print(f\"\\nFirst few trading signals:\")\n", + " for ii, (idx, trade) in enumerate(pair_trades.head(NTRADES_TO_SHOW).iterrows()):\n", + " print(f\" {ii+1}. {trade['action']} {trade['symbol']} @ ${trade['price']:.2f} at {trade['time']}\")\n", + " \n", + " if len(pair_trades) > NTRADES_TO_SHOW:\n", + " print(f\" ... and {len(pair_trades) - NTRADES_TO_SHOW} more signals\")\n", + " \n", + " else:\n", + " print(f\"\\nTrading Signals: None generated\")\n", + " print(\" Possible reasons:\")\n", + " print(\" - Dis-equilibrium never exceeded open threshold\")\n", + " print(\" - Pair not cointegrated (for StaticFit)\")\n", + " print(\" - Insufficient data or market conditions\")\n", + "\n", + " print(f\"\\n\" + \"=\" * 80)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Performance" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "metadata": {}, + "outputs": [], + "source": [ + "def performance_results() -> None:\n", + " from datetime import datetime\n", + "\n", + " if pair_trades is not None and len(pair_trades) > 0:\n", + " # Print detailed results using BacktestResult methods\n", + " bt_result.print_single_day_results()\n", + " \n", + " # Print trading signal details\n", + " print(f\"\\nDetailed Trading Signals:\")\n", + " print(f\"{'Time':<20} {'Action':<15} {'Symbol':<10} {'Price':<12} {'Scaled Dis-eq':<15} {'Status':<10}\")\n", + " print(\"-\" * 90)\n", + " \n", + " for _, trade in pair_trades.head(10).iterrows(): # Show first 10 trades\n", + " time_str = str(trade['time'])[:19] \n", + " action_str = str(trade['action'])[:14]\n", + " symbol_str = str(trade['symbol'])[:9]\n", + " price_str = f\"${trade['price']:.2f}\"\n", + " diseq_str = f\"{trade.get('scaled_disequilibrium', 'N/A'):.3f}\" if 'scaled_disequilibrium' in trade else 'N/A'\n", + " status = trade.get('status', 'N/A')\n", + " \n", + " print(f\"{time_str:<20} {action_str:<15} {symbol_str:<10} {price_str:<12} {diseq_str:<15} {status:<10}\")\n", + " \n", + " if len(pair_trades) > 10:\n", + " print(f\"... and {len(pair_trades)-10} more trading signals\")\n", + " \n", + " # Print outstanding positions\n", + " bt_result.print_outstanding_positions()\n", + " \n", + " # Print grand totals\n", + " bt_result.print_grand_totals()\n", + " \n", + " else:\n", + " print(f\"\\nNo trading signals generated\")\n", + " print(f\"Backtest completed with no trades\")\n", + " \n", + " # Still print any outstanding information\n", + " print(f\"\\nConfiguration Summary:\")\n", + " print(f\" Pair: {SYMBOL_A} & {SYMBOL_B}\")\n", + " print(f\" Strategy: {FIT_METHOD_TYPE}\")\n", + " print(f\" Open threshold: {PT_BT_CONFIG['dis-equilibrium_open_trshld']}\")\n", + " print(f\" Close threshold: {pt_bt_config['dis-equilibrium_close_trshld']}\")\n", + " print(f\" Training window: {pt_bt_config['training_minutes']} minutes\")\n", + " \n", + " print(\"\\n\" + \"=\"*80)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Run" + ] + }, + { + "cell_type": "code", + "execution_count": 24, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ + "Setup complete!\n", "Trading Parameters:\n", " Configuration: equity\n", " Symbol A: COIN\n", @@ -235,92 +1299,7 @@ "Data Configuration:\n", " Data File: 20250604.mktdata.ohlcv.db\n", " Security Type: EQUITY\n", - " ✓ Data file found: /home/oleg/develop/pairs_trading/data/equity/20250604.mktdata.ohlcv.db\n" - ] - } - ], - "source": [ - "print(f\"Trading Parameters:\")\n", - "print(f\" Configuration: {CONFIG_FILE}\")\n", - "print(f\" Symbol A: {SYMBOL_A}\")\n", - "print(f\" Symbol B: {SYMBOL_B}\")\n", - "print(f\" Trading Date: {TRD_DATE}\")\n", - "\n", - "# Load the specified configuration\n", - "print(f\"\\nLoading {CONFIG_FILE} configuration using HJSON...\")\n", - "\n", - "CONFIG = load_config_from_file(CONFIG_FILE)\n", - "assert CONFIG is not None\n", - "pt_bt_config: Dict = dict(CONFIG)\n", - "\n", - "if pt_bt_config:\n", - " print(f\"✓ Successfully loaded {pt_bt_config['security_type']} configuration\")\n", - " print(f\" Data directory: {pt_bt_config['data_directory']}\")\n", - " print(f\" Database table: {pt_bt_config['db_table_name']}\")\n", - " print(f\" Exchange: {pt_bt_config['exchange_id']}\")\n", - " print(f\" Training window: {pt_bt_config['training_minutes']} minutes\")\n", - " print(f\" Open threshold: {pt_bt_config['dis-equilibrium_open_trshld']}\")\n", - " print(f\" Close threshold: {pt_bt_config['dis-equilibrium_close_trshld']}\")\n", - " \n", - " # Instantiate strategy from config\n", - " FIT_MODEL = instantiate_fit_method_from_config(pt_bt_config)\n", - " print(f\" Fit Method: {type(FIT_MODEL).__name__}\")\n", - " \n", - " # Automatically construct data file name based on date and config type\n", - " DATA_FILE = f\"{TRADING_DATE}.mktdata.ohlcv.db\"\n", - "\n", - " # Update CONFIG with the specific data file and instruments\n", - " pt_bt_config[\"datafiles\"] = [DATA_FILE]\n", - " pt_bt_config[\"instruments\"] = [SYMBOL_A, SYMBOL_B]\n", - " \n", - " print(f\"\\nData Configuration:\")\n", - " print(f\" Data File: {DATA_FILE}\")\n", - " print(f\" Security Type: {pt_bt_config['security_type']}\")\n", - " \n", - " # Verify data file exists\n", - " data_file_path = f\"{pt_bt_config['data_directory']}/{DATA_FILE}\"\n", - " if os.path.exists(data_file_path):\n", - " print(f\" ✓ Data file found: {data_file_path}\")\n", - " else:\n", - " print(f\" ⚠ Data file not found: {data_file_path}\")\n", - " print(f\" Please check if the date and file exist in the data directory\")\n", - " \n", - " # List available files in the data directory\n", - " try:\n", - " data_dir = pt_bt_config['data_directory']\n", - " if os.path.exists(data_dir):\n", - " available_files = [f for f in os.listdir(data_dir) if f.endswith('.db')]\n", - " print(f\" Available files in {data_dir}:\")\n", - " for file in sorted(available_files)[:5]: # Show first 5 files\n", - " print(f\" - {file}\")\n", - " if len(available_files) > 5:\n", - " print(f\" ... and {len(available_files)-5} more files\")\n", - " except Exception as e:\n", - " print(f\" Could not list files in data directory: {e}\")\n", - "else:\n", - " print(\"⚠ Failed to load configuration. Please check the configuration file.\")\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "vscode": { - "languageId": "raw" - } - }, - "source": [ - "## Load and Prepare Market Data\n" - ] - }, - { - "cell_type": "code", - "execution_count": 51, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ + " ✓ Data file found: /home/oleg/develop/pairs_trading/data/equity/20250604.mktdata.ohlcv.db\n", "Loading data from: /home/oleg/develop/pairs_trading/data/equity/20250604.mktdata.ohlcv.db\n", "Loaded 781 rows of market data\n", "Symbols in data: ['COIN' 'MSTR']\n", @@ -405,59 +1384,12 @@ }, "metadata": {}, "output_type": "display_data" - } - ], - "source": [ - "# Load market data\n", - "datafile_path = f\"{pt_bt_config['data_directory']}/{DATA_FILE}\"\n", - "print(f\"Loading data from: {datafile_path}\")\n", - "\n", - "market_data_df = load_market_data(datafile_path, config=pt_bt_config)\n", - "\n", - "print(f\"Loaded {len(market_data_df)} rows of market data\")\n", - "print(f\"Symbols in data: {market_data_df['symbol'].unique()}\")\n", - "print(f\"Time range: {market_data_df['tstamp'].min()} to {market_data_df['tstamp'].max()}\")\n", - "\n", - "# Create trading pair\n", - "pair = TradingPair(\n", - " market_data=market_data_df,\n", - " symbol_a=SYMBOL_A,\n", - " symbol_b=SYMBOL_B,\n", - " price_column=pt_bt_config[\"price_column\"]\n", - ")\n", - "\n", - "print(f\"\\nCreated trading pair: {pair}\")\n", - "print(f\"Market data shape: {pair.market_data_.shape}\")\n", - "print(f\"Column names: {pair.colnames()}\")\n", - "\n", - "# Display sample data\n", - "print(f\"\\nSample data:\")\n", - "display(pair.market_data_.head())\n", - "\n", - "TRADING_DATE = f\"{pair.market_data_['tstamp'].min()} to {pair.market_data_['tstamp'].max()}\"" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "vscode": { - "languageId": "raw" - } - }, - "source": [ - "## Print Strategy Specifics\n" - ] - }, - { - "cell_type": "code", - "execution_count": 52, - "metadata": {}, - "outputs": [ + }, { "name": "stdout", "output_type": "stream", "text": [ - "Analysis for SlidingFit...\n", + "Analysis for SlidingFit ...\n", "\n", "=== SLIDING FIT FIT_MODEL ANALYSIS ===\n", "This strategy:\n", @@ -476,50 +1408,7 @@ " Training minutes: 120\n", " Funding per pair: $2000\n" ] - } - ], - "source": [ - "# Determine analysis approach based on strategy type\n", - "FIT_METHOD_TYPE = type(FIT_MODEL).__name__\n", - "print(f\"Analysis for {FIT_METHOD_TYPE}...\")\n", - "\n", - "print(\"\\n=== SLIDING FIT FIT_MODEL ANALYSIS ===\")\n", - "print(\"This strategy:\")\n", - "print(\" - Re-fits cointegration model using sliding window\")\n", - "print(\" - Adapts to changing market conditions\")\n", - "print(\" - Dynamic parameter updates every minute\")\n", - "\n", - "# Calculate maximum possible iterations for sliding window\n", - "training_minutes = pt_bt_config[\"training_minutes\"]\n", - "max_iterations = len(pair.market_data_) - training_minutes\n", - "print(f\"\\nSliding window analysis parameters:\")\n", - "print(f\" Training window size: {training_minutes} minutes\")\n", - "print(f\" Maximum iterations: {max_iterations}\")\n", - "print(f\" Total analysis time: ~{max_iterations} minutes\")\n", - "\n", - "print(f\"\\nStrategy Configuration:\")\n", - "print(f\" Open threshold: {pt_bt_config['dis-equilibrium_open_trshld']}\")\n", - "print(f\" Close threshold: {pt_bt_config['dis-equilibrium_close_trshld']}\")\n", - "print(f\" Training minutes: {pt_bt_config['training_minutes']}\")\n", - "print(f\" Funding per pair: ${pt_bt_config['funding_per_pair']}\")\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "vscode": { - "languageId": "raw" - } - }, - "source": [ - "## Visualize Raw Price Data\n" - ] - }, - { - "cell_type": "code", - "execution_count": 53, - "metadata": {}, - "outputs": [ + }, { "data": { "image/png": "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", @@ -549,106 +1438,7 @@ " COIN: Mean=$256.28, Std=$1.13\n", " MSTR: Mean=$382.71, Std=$2.30\n", " Price Ratio: Mean=0.67, Std=0.01\n", - " Correlation: -0.0564\n" - ] - } - ], - "source": [ - "# Plot raw price data\n", - "\n", - "# Get column names for the trading pair\n", - "colname_a, colname_b = pair.colnames()\n", - "price_data = pair.market_data_.copy()\n", - "\n", - "# # 1. Price data - separate plots for each symbol\n", - "# colname_a, colname_b = pair.colnames()\n", - "# price_data = pair.market_data_.copy()\n", - "\n", - "# Create separate subplots for better visibility\n", - "fig_price, price_axes = plt.subplots(2, 1, figsize=(18, 10))\n", - "\n", - "# Plot SYMBOL_A\n", - "price_axes[0].plot(price_data['tstamp'], price_data[colname_a], alpha=0.7, \n", - " label=f'{SYMBOL_A}', linewidth=1, color='blue')\n", - "price_axes[0].set_title(f'{SYMBOL_A} Price Data ({TRD_DATE})')\n", - "price_axes[0].set_ylabel(f'{SYMBOL_A} Price')\n", - "price_axes[0].legend()\n", - "price_axes[0].grid(True)\n", - "\n", - "# Plot SYMBOL_B\n", - "price_axes[1].plot(price_data['tstamp'], price_data[colname_b], alpha=0.7, \n", - " label=f'{SYMBOL_B}', linewidth=1, color='red')\n", - "price_axes[1].set_title(f'{SYMBOL_B} Price Data ({TRD_DATE})')\n", - "price_axes[1].set_ylabel(f'{SYMBOL_B} Price')\n", - "price_axes[1].set_xlabel('Time')\n", - "price_axes[1].legend()\n", - "price_axes[1].grid(True)\n", - "\n", - "plt.tight_layout()\n", - "plt.show()\n", - " \n", - "\n", - "# Plot individual prices\n", - "fig, axes = plt.subplots(2, 1, figsize=(18, 12))\n", - "\n", - "# Normalized prices for comparison\n", - "norm_a = price_data[colname_a] / price_data[colname_a].iloc[0]\n", - "norm_b = price_data[colname_b] / price_data[colname_b].iloc[0]\n", - "\n", - "axes[0].plot(price_data['tstamp'], norm_a, label=f'{SYMBOL_A} (normalized)', alpha=0.8, linewidth=1)\n", - "axes[0].plot(price_data['tstamp'], norm_b, label=f'{SYMBOL_B} (normalized)', alpha=0.8, linewidth=1)\n", - "axes[0].set_title(f'Normalized Price Comparison (Base = 1.0) ({TRD_DATE})')\n", - "axes[0].set_ylabel('Normalized Price')\n", - "axes[0].legend()\n", - "axes[0].grid(True)\n", - "\n", - "# Price ratio\n", - "price_ratio = price_data[colname_a] / price_data[colname_b]\n", - "axes[1].plot(price_data['tstamp'], price_ratio, label=f'{SYMBOL_A}/{SYMBOL_B} Ratio', color='green', alpha=0.8, linewidth=1)\n", - "axes[1].set_title(f'Price Ratio Px({SYMBOL_A})/Px({SYMBOL_B}) ({TRD_DATE})')\n", - "axes[1].set_ylabel('Ratio')\n", - "axes[1].set_xlabel('Time')\n", - "axes[1].legend()\n", - "axes[1].grid(True)\n", - "\n", - "plt.tight_layout()\n", - "plt.show()\n", - "\n", - "# Print basic statistics\n", - "print(f\"\\nPrice Statistics:\")\n", - "print(f\" {SYMBOL_A}: Mean=${price_data[colname_a].mean():.2f}, Std=${price_data[colname_a].std():.2f}\")\n", - "print(f\" {SYMBOL_B}: Mean=${price_data[colname_b].mean():.2f}, Std=${price_data[colname_b].std():.2f}\")\n", - "print(f\" Price Ratio: Mean={price_ratio.mean():.2f}, Std={price_ratio.std():.2f}\")\n", - "print(f\" Correlation: {price_data[colname_a].corr(price_data[colname_b]):.4f}\")\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "vscode": { - "languageId": "raw" - } - }, - "source": [ - "# Run" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Analysis" - ] - }, - { - "cell_type": "code", - "execution_count": 54, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ + " Correlation: -0.0564\n", "Running SlidingFit analysis...\n", "\n", "=== SLIDING FIT ANALYSIS ===\n", @@ -657,13 +1447,6 @@ "********************************************************************************\n", "Pair COIN & MSTR (0) IS COINTEGRATED\n", "********************************************************************************\n", - "265\r" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ "COIN & MSTR: current offset=271 * Training data length=119 < 120 * Not enough training data. Completing the job.\n", "OPEN_TRADES: 2025-06-04 15:33:00 open_scaled_disequilibrium=np.float64(2.2136223219159255)\n", "OPEN TRADES:\n", @@ -724,68 +1507,7 @@ "BACKTEST RESULTS\n", "================================================================================\n" ] - } - ], - "source": [ - "# Initialize strategy state and run analysis\n", - "print(f\"Running {FIT_METHOD_TYPE} analysis...\")\n", - "\n", - "# Initialize result tracking\n", - "bt_result = BacktestResult(config=pt_bt_config)\n", - "pair_trades = None\n", - "\n", - "# Run strategy-specific analysis\n", - "print(\"\\n=== SLIDING FIT ANALYSIS ===\")\n", - "\n", - "# Initialize tracking variables for sliding window analysis\n", - "training_minutes = pt_bt_config[\"training_minutes\"]\n", - "max_iterations = len(pair.market_data_) - training_minutes\n", - "\n", - "# Limit iterations for demonstration (change this for full run)\n", - "max_demo_iterations = min(200, max_iterations)\n", - "print(f\"Processing first {max_demo_iterations} iterations for demonstration...\")\n", - "\n", - "# Initialize pair state for sliding fit method\n", - "pair.user_data_['state'] = PairState.INITIAL\n", - "pair.user_data_[\"trades\"] = pd.DataFrame(columns=pd.Index(FIT_MODEL.TRADES_COLUMNS, dtype=str))\n", - "pair.user_data_[\"is_cointegrated\"] = False\n", - "\n", - "# Run the sliding fit method\n", - "# ==========================================================================\n", - "pair_trades = FIT_MODEL.run_pair(config=pt_bt_config, pair=pair, bt_result=bt_result)\n", - "# ==========================================================================\n", - "\n", - "if pair_trades is not None and len(pair_trades) > 0:\n", - " print(f\"Generated {len(pair_trades)} trading signals\")\n", - "else:\n", - " print(\"No trading signals generated\")\n", - "\n", - "print(\"\\nStrategy execution completed!\")\n", - "\n", - "# Print comprehensive backtest results\n", - "print(\"\\n\" + \"=\"*80)\n", - "print(\"BACKTEST RESULTS\")\n", - "print(\"=\"*80)\n", - "\n", - "assert pair.predicted_df_ is not None\n" - ] - }, - { - "cell_type": "markdown", - "metadata": { - "vscode": { - "languageId": "raw" - } - }, - "source": [ - "## Visualization\n" - ] - }, - { - "cell_type": "code", - "execution_count": 55, - "metadata": {}, - "outputs": [ + }, { "data": { "text/html": [ @@ -3266,9 +3988,9 @@ }, "text/html": [ "
\n", - "
\n", - "