nb-cli 0.0.6

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "249c2bd5-8dab-45fe-bb5f-07cd98b91576",
   "metadata": {},
   "source": [
    "# USO Stock Price Forecasting with Chronos-2\n",
    "\n",
    "This notebook forecasts USO stock prices for the next 30 days using the Chronos-2 model with GPU acceleration support.\n",
    "\n",
    "**GPU Support:**\n",
    "- ✅ **MPS** (Apple Silicon M1/M2/M3/M4)\n",
    "- ✅ **CUDA** (NVIDIA GPUs)\n",
    "- ✅ **CPU** fallback (if no GPU available)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "babdb521-9713-4edb-a071-3e82b4604d9d",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ['KMP_DUPLICATE_LIB_OK'] = 'TRUE'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "79b85034-7a2a-4c58-9d15-caa2184aea07",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "45a3f38a-a46e-440f-afdc-5647f6c63e87",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "✅ Using MPS (Metal Performance Shaders) for GPU acceleration\n",
      "Device: mps\n"
     ]
    }
   ],
   "source": [
    "# Setup device (MPS for Apple Silicon, CUDA for NVIDIA GPUs, or CPU)\n",
    "import torch\n",
    "\n",
    "if torch.backends.mps.is_available():\n",
    "    device = \"mps\"\n",
    "    print(\"✅ Using MPS (Metal Performance Shaders) for GPU acceleration\")\n",
    "elif torch.cuda.is_available():\n",
    "    device = \"cuda\"\n",
    "    print(\"✅ Using CUDA for GPU acceleration\")\n",
    "else:\n",
    "    device = \"cpu\"\n",
    "    print(\"⚠️ Using CPU (no GPU acceleration available)\")\n",
    "\n",
    "print(f\"Device: {device}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b05dfcf3-bd42-4ed5-9c4c-59972874a324",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Install required packages\n",
    "!pip install -q \"chronos-forecasting>=2.0\" yfinance pandas matplotlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "cad7b067-49c6-485e-b173-8c373caf9f96",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import yfinance as yf\n",
    "from datetime import datetime, timedelta\n",
    "from chronos import Chronos2Pipeline"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c1461a7-ef40-4590-8ba9-56fe6e221c19",
   "metadata": {},
   "source": [
    "## Fetch Historical Stock Data\n",
    "\n",
    "Downloading the past year of USO stock data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "55aac89f-b68a-40dc-8e26-bd0a7ff69822",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fetching USO stock data from 2025-03-16 to 2026-03-16\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\nDownloaded 250 trading days of data\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Volume</th>\n",
       "      <th>Dividends</th>\n",
       "      <th>Stock Splits</th>\n",
       "      <th>Capital Gains</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2025-03-17 00:00:00-04:00</th>\n",
       "      <td>72.980003</td>\n",
       "      <td>73.169998</td>\n",
       "      <td>72.430000</td>\n",
       "      <td>72.709999</td>\n",
       "      <td>2873700</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2025-03-18 00:00:00-04:00</th>\n",
       "      <td>73.389999</td>\n",
       "      <td>73.489998</td>\n",
       "      <td>71.970001</td>\n",
       "      <td>72.209999</td>\n",
       "      <td>2739000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2025-03-19 00:00:00-04:00</th>\n",
       "      <td>72.080002</td>\n",
       "      <td>72.879997</td>\n",
       "      <td>71.930000</td>\n",
       "      <td>72.419998</td>\n",
       "      <td>1747400</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2025-03-20 00:00:00-04:00</th>\n",
       "      <td>72.230003</td>\n",
       "      <td>73.800003</td>\n",
       "      <td>72.040001</td>\n",
       "      <td>73.730003</td>\n",
       "      <td>2523100</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2025-03-21 00:00:00-04:00</th>\n",
       "      <td>73.389999</td>\n",
       "      <td>73.959999</td>\n",
       "      <td>73.139999</td>\n",
       "      <td>73.790001</td>\n",
       "      <td>1528800</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                Open       High        Low      Close  \\\n",
       "Date                                                                    \n",
       "2025-03-17 00:00:00-04:00  72.980003  73.169998  72.430000  72.709999   \n",
       "2025-03-18 00:00:00-04:00  73.389999  73.489998  71.970001  72.209999   \n",
       "2025-03-19 00:00:00-04:00  72.080002  72.879997  71.930000  72.419998   \n",
       "2025-03-20 00:00:00-04:00  72.230003  73.800003  72.040001  73.730003   \n",
       "2025-03-21 00:00:00-04:00  73.389999  73.959999  73.139999  73.790001   \n",
       "\n",
       "                            Volume  Dividends  Stock Splits  Capital Gains  \n",
       "Date                                                                        \n",
       "2025-03-17 00:00:00-04:00  2873700        0.0           0.0            0.0  \n",
       "2025-03-18 00:00:00-04:00  2739000        0.0           0.0            0.0  \n",
       "2025-03-19 00:00:00-04:00  1747400        0.0           0.0            0.0  \n",
       "2025-03-20 00:00:00-04:00  2523100        0.0           0.0            0.0  \n",
       "2025-03-21 00:00:00-04:00  1528800        0.0           0.0            0.0  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define date range (past 1 year)\n",
    "end_date = datetime(2026, 3, 16)  # Today's date\n",
    "start_date = end_date - timedelta(days=365)\n",
    "\n",
    "print(f\"Fetching USO stock data from {start_date.date()} to {end_date.date()}\")\n",
    "\n",
    "# Download stock data\n",
    "ticker = yf.Ticker(\"USO\")\n",
    "stock_data = ticker.history(start=start_date, end=end_date)\n",
    "\n",
    "print(f\"\\\\nDownloaded {len(stock_data)} trading days of data\")\n",
    "stock_data.head()\n",
    ""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "709b222e-0b0e-4a0f-825a-31412e73c438",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prepared 260 data points for forecasting\n",
      "\\nDate range: 2025-03-17 00:00:00-04:00 to 2026-03-13 00:00:00-04:00\n",
      "Price range: $62.37 to $119.89\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>timestamp</th>\n",
       "      <th>target</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>255</th>\n",
       "      <td>2026-03-09 00:00:00-04:00</td>\n",
       "      <td>104.330002</td>\n",
       "      <td>USO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>256</th>\n",
       "      <td>2026-03-10 00:00:00-04:00</td>\n",
       "      <td>105.860001</td>\n",
       "      <td>USO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>257</th>\n",
       "      <td>2026-03-11 00:00:00-04:00</td>\n",
       "      <td>108.050003</td>\n",
       "      <td>USO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>258</th>\n",
       "      <td>2026-03-12 00:00:00-04:00</td>\n",
       "      <td>118.389999</td>\n",
       "      <td>USO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>259</th>\n",
       "      <td>2026-03-13 00:00:00-04:00</td>\n",
       "      <td>119.889999</td>\n",
       "      <td>USO</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    timestamp      target   id\n",
       "255 2026-03-09 00:00:00-04:00  104.330002  USO\n",
       "256 2026-03-10 00:00:00-04:00  105.860001  USO\n",
       "257 2026-03-11 00:00:00-04:00  108.050003  USO\n",
       "258 2026-03-12 00:00:00-04:00  118.389999  USO\n",
       "259 2026-03-13 00:00:00-04:00  119.889999  USO"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Prepare data for Chronos-2\n",
    "# We'll use the closing price for forecasting\n",
    "\n",
    "# Resample to business days and forward fill to ensure consistent frequency\n",
    "stock_data_resampled = stock_data.resample('B').ffill()\n",
    "\n",
    "context_df = pd.DataFrame({\n",
    "    'timestamp': stock_data_resampled.index,\n",
    "    'target': stock_data_resampled['Close'].values,\n",
    "    'id': 'USO'\n",
    "})\n",
    "\n",
    "context_df = context_df.reset_index(drop=True)\n",
    "print(f\"Prepared {len(context_df)} data points for forecasting\")\n",
    "print(f\"\\\\nDate range: {context_df['timestamp'].min()} to {context_df['timestamp'].max()}\")\n",
    "print(f\"Price range: ${context_df['target'].min():.2f} to ${context_df['target'].max():.2f}\")\n",
    "context_df.tail()\n",
    ""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e54cb875-4669-45e1-9e79-f66e8fcb0793",
   "metadata": {},
   "source": [
    "## Load Chronos-2 Model and Generate Forecast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7b5215d1-6f67-4635-a235-814d2b670e73",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading Chronos-2 model...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model loaded successfully!\n"
     ]
    }
   ],
   "source": [
    "# Load Chronos-2 model with automatic device detection\n",
    "print(\"Loading Chronos-2 model...\")\n",
    "pipeline = Chronos2Pipeline.from_pretrained(\"amazon/chronos-2\", device_map=device)\n",
    "print(\"Model loaded successfully!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "47ea4b6d-63da-45ff-b08a-0c07724e4b99",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Generating 30-day forecast for USO...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Forecast complete!\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>target_name</th>\n",
       "      <th>predictions</th>\n",
       "      <th>0.1</th>\n",
       "      <th>0.5</th>\n",
       "      <th>0.9</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>USO</td>\n",
       "      <td>2026-03-16 00:00:00-04:00</td>\n",
       "      <td>target</td>\n",
       "      <td>118.323624</td>\n",
       "      <td>109.486938</td>\n",
       "      <td>118.323624</td>\n",
       "      <td>129.094040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>USO</td>\n",
       "      <td>2026-03-17 00:00:00-04:00</td>\n",
       "      <td>target</td>\n",
       "      <td>117.568039</td>\n",
       "      <td>105.867233</td>\n",
       "      <td>117.568039</td>\n",
       "      <td>128.783691</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>USO</td>\n",
       "      <td>2026-03-18 00:00:00-04:00</td>\n",
       "      <td>target</td>\n",
       "      <td>117.541130</td>\n",
       "      <td>103.975113</td>\n",
       "      <td>117.541130</td>\n",
       "      <td>131.241852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>USO</td>\n",
       "      <td>2026-03-19 00:00:00-04:00</td>\n",
       "      <td>target</td>\n",
       "      <td>117.148743</td>\n",
       "      <td>99.623070</td>\n",
       "      <td>117.148743</td>\n",
       "      <td>133.644791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>USO</td>\n",
       "      <td>2026-03-20 00:00:00-04:00</td>\n",
       "      <td>target</td>\n",
       "      <td>116.921593</td>\n",
       "      <td>97.649628</td>\n",
       "      <td>116.921593</td>\n",
       "      <td>135.024124</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    id                 timestamp target_name  predictions         0.1  \\\n",
       "0  USO 2026-03-16 00:00:00-04:00      target   118.323624  109.486938   \n",
       "1  USO 2026-03-17 00:00:00-04:00      target   117.568039  105.867233   \n",
       "2  USO 2026-03-18 00:00:00-04:00      target   117.541130  103.975113   \n",
       "3  USO 2026-03-19 00:00:00-04:00      target   117.148743   99.623070   \n",
       "4  USO 2026-03-20 00:00:00-04:00      target   116.921593   97.649628   \n",
       "\n",
       "          0.5         0.9  \n",
       "0  118.323624  129.094040  \n",
       "1  117.568039  128.783691  \n",
       "2  117.541130  131.241852  \n",
       "3  117.148743  133.644791  \n",
       "4  116.921593  135.024124  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generate 30-day forecast\n",
    "prediction_length = 30  # 30 trading days\n",
    "\n",
    "print(f\"Generating {prediction_length}-day forecast for USO...\")\n",
    "\n",
    "pred_df = pipeline.predict_df(\n",
    "    context_df,\n",
    "    prediction_length=prediction_length,\n",
    "    quantile_levels=[0.1, 0.5, 0.9],  # 10th, 50th (median), and 90th percentiles\n",
    "    id_column=\"id\",\n",
    "    timestamp_column=\"timestamp\",\n",
    "    target=\"target\",\n",
    ")\n",
    "\n",
    "print(\"Forecast complete!\")\n",
    "pred_df.head()\n",
    ""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "99a62cbb-d7a6-4720-a707-a87e69bd7dff",
   "metadata": {},
   "source": [
    "## Visualize Forecast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "2e171c64-038c-4f73-9945-8a9b25beceaf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Show the last 60 days of historical data for context\n",
    "ts_context = context_df.set_index(\"timestamp\")[\"target\"].tail(60)\n",
    "ts_pred = pred_df.set_index(\"timestamp\")\n",
    "\n",
    "plt.figure(figsize=(14, 6))\n",
    "ts_context.plot(label=\"Historical Data (Last 60 Days)\", color=\"#1f77b4\", linewidth=2)\n",
    "ts_pred[\"predictions\"].plot(label=\"30-Day Forecast (Median)\", color=\"#ff7f0e\", linewidth=2)\n",
    "\n",
    "# Add prediction interval (10th to 90th percentile)\n",
    "plt.fill_between(\n",
    "    ts_pred.index,\n",
    "    ts_pred[\"0.1\"],\n",
    "    ts_pred[\"0.9\"],\n",
    "    alpha=0.3,\n",
    "    label=\"80% Prediction Interval (10th-90th percentile)\",\n",
    "    color=\"#ff7f0e\",\n",
    ")\n",
    "\n",
    "plt.title(\"USO Stock Price Forecast - Next 30 Days\", fontsize=16, fontweight='bold')\n",
    "plt.xlabel(\"Date\", fontsize=12)\n",
    "plt.ylabel(\"Price ($)\", fontsize=12)\n",
    "plt.legend(loc='best', fontsize=10)\n",
    "plt.grid(True, alpha=0.3)\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ddbbb54e-cd8b-4991-b63f-1e4d5f1c7849",
   "metadata": {},
   "source": [
    "## Forecast Summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e29e8a14-a720-4d5c-b3cb-cf64c46543ab",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "============================================================\n",
      "USO STOCK FORECAST SUMMARY\n",
      "============================================================\n",
      "\\nCurrent Price (as of 2026-03-13): $119.89\n",
      "\\nForecast for 2026-04-24 (30 days ahead):\n",
      "  Median Prediction: $111.56\n",
      "  10th Percentile:   $70.56\n",
      "  90th Percentile:   $156.64\n",
      "\\nExpected Change: $-8.33 (-6.95%)\n",
      "\\nNote: This is a probabilistic forecast. The 80% prediction interval\n",
      "suggests the price is likely to fall between the 10th and 90th percentiles.\n",
      "============================================================\n"
     ]
    }
   ],
   "source": [
    "# Print forecast summary\n",
    "last_price = context_df['target'].iloc[-1]\n",
    "forecast_median = ts_pred['predictions'].iloc[-1]\n",
    "forecast_low = ts_pred['0.1'].iloc[-1]\n",
    "forecast_high = ts_pred['0.9'].iloc[-1]\n",
    "\n",
    "print(\"=\"*60)\n",
    "print(\"USO STOCK FORECAST SUMMARY\")\n",
    "print(\"=\"*60)\n",
    "print(f\"\\\\nCurrent Price (as of {context_df['timestamp'].iloc[-1].date()}): ${last_price:.2f}\")\n",
    "print(f\"\\\\nForecast for {pred_df['timestamp'].iloc[-1].date()} (30 days ahead):\")\n",
    "print(f\"  Median Prediction: ${forecast_median:.2f}\")\n",
    "print(f\"  10th Percentile:   ${forecast_low:.2f}\")\n",
    "print(f\"  90th Percentile:   ${forecast_high:.2f}\")\n",
    "print(f\"\\\\nExpected Change: ${forecast_median - last_price:.2f} ({((forecast_median - last_price) / last_price * 100):.2f}%)\")\n",
    "print(\"\\\\nNote: This is a probabilistic forecast. The 80% prediction interval\")\n",
    "print(\"suggests the price is likely to fall between the 10th and 90th percentiles.\")\n",
    "print(\"=\"*60)\n",
    ""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d3153e8f-f319-42bc-a56b-5b1b4411be6c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\nDetailed 30-Day Forecast:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Lower Bound (10%)</th>\n",
       "      <th>Median Forecast</th>\n",
       "      <th>Upper Bound (90%)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2026-03-16 00:00:00-04:00</th>\n",
       "      <td>109.489998</td>\n",
       "      <td>118.320000</td>\n",
       "      <td>129.089996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-03-17 00:00:00-04:00</th>\n",
       "      <td>105.870003</td>\n",
       "      <td>117.570000</td>\n",
       "      <td>128.779999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-03-18 00:00:00-04:00</th>\n",
       "      <td>103.980003</td>\n",
       "      <td>117.540001</td>\n",
       "      <td>131.240005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-03-19 00:00:00-04:00</th>\n",
       "      <td>99.620003</td>\n",
       "      <td>117.150002</td>\n",
       "      <td>133.639999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-03-20 00:00:00-04:00</th>\n",
       "      <td>97.650002</td>\n",
       "      <td>116.919998</td>\n",
       "      <td>135.020004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-03-23 00:00:00-04:00</th>\n",
       "      <td>94.940002</td>\n",
       "      <td>116.349998</td>\n",
       "      <td>136.009995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-03-24 00:00:00-04:00</th>\n",
       "      <td>91.430000</td>\n",
       "      <td>116.529999</td>\n",
       "      <td>138.419998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-03-25 00:00:00-04:00</th>\n",
       "      <td>89.970001</td>\n",
       "      <td>115.430000</td>\n",
       "      <td>138.869995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-03-26 00:00:00-04:00</th>\n",
       "      <td>88.519997</td>\n",
       "      <td>115.059998</td>\n",
       "      <td>140.570007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-03-27 00:00:00-04:00</th>\n",
       "      <td>86.169998</td>\n",
       "      <td>115.099998</td>\n",
       "      <td>140.649994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-03-30 00:00:00-04:00</th>\n",
       "      <td>84.739998</td>\n",
       "      <td>114.980003</td>\n",
       "      <td>143.369995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-03-31 00:00:00-04:00</th>\n",
       "      <td>83.000000</td>\n",
       "      <td>114.459999</td>\n",
       "      <td>145.669998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-01 00:00:00-04:00</th>\n",
       "      <td>81.940002</td>\n",
       "      <td>114.209999</td>\n",
       "      <td>145.270004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-02 00:00:00-04:00</th>\n",
       "      <td>80.379997</td>\n",
       "      <td>114.809998</td>\n",
       "      <td>146.860001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-03 00:00:00-04:00</th>\n",
       "      <td>80.269997</td>\n",
       "      <td>115.389999</td>\n",
       "      <td>148.309998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-06 00:00:00-04:00</th>\n",
       "      <td>79.019997</td>\n",
       "      <td>114.830002</td>\n",
       "      <td>150.889999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-07 00:00:00-04:00</th>\n",
       "      <td>80.900002</td>\n",
       "      <td>116.309998</td>\n",
       "      <td>147.710007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-08 00:00:00-04:00</th>\n",
       "      <td>79.139999</td>\n",
       "      <td>116.779999</td>\n",
       "      <td>148.910004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-09 00:00:00-04:00</th>\n",
       "      <td>78.669998</td>\n",
       "      <td>117.029999</td>\n",
       "      <td>148.729996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-10 00:00:00-04:00</th>\n",
       "      <td>78.650002</td>\n",
       "      <td>117.059998</td>\n",
       "      <td>149.539993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-13 00:00:00-04:00</th>\n",
       "      <td>78.070000</td>\n",
       "      <td>117.790001</td>\n",
       "      <td>150.850006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-14 00:00:00-04:00</th>\n",
       "      <td>77.400002</td>\n",
       "      <td>116.650002</td>\n",
       "      <td>150.470001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-15 00:00:00-04:00</th>\n",
       "      <td>75.739998</td>\n",
       "      <td>116.559998</td>\n",
       "      <td>151.869995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-16 00:00:00-04:00</th>\n",
       "      <td>75.209999</td>\n",
       "      <td>115.980003</td>\n",
       "      <td>152.679993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-17 00:00:00-04:00</th>\n",
       "      <td>74.919998</td>\n",
       "      <td>115.320000</td>\n",
       "      <td>154.320007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-20 00:00:00-04:00</th>\n",
       "      <td>73.400002</td>\n",
       "      <td>115.290001</td>\n",
       "      <td>153.389999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-21 00:00:00-04:00</th>\n",
       "      <td>72.790001</td>\n",
       "      <td>114.139999</td>\n",
       "      <td>155.850006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-22 00:00:00-04:00</th>\n",
       "      <td>72.199997</td>\n",
       "      <td>112.820000</td>\n",
       "      <td>157.600006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-23 00:00:00-04:00</th>\n",
       "      <td>71.629997</td>\n",
       "      <td>111.230003</td>\n",
       "      <td>156.720001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-04-24 00:00:00-04:00</th>\n",
       "      <td>70.559998</td>\n",
       "      <td>111.559998</td>\n",
       "      <td>156.639999</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           Lower Bound (10%)  Median Forecast  \\\n",
       "Date                                                            \n",
       "2026-03-16 00:00:00-04:00         109.489998       118.320000   \n",
       "2026-03-17 00:00:00-04:00         105.870003       117.570000   \n",
       "2026-03-18 00:00:00-04:00         103.980003       117.540001   \n",
       "2026-03-19 00:00:00-04:00          99.620003       117.150002   \n",
       "2026-03-20 00:00:00-04:00          97.650002       116.919998   \n",
       "2026-03-23 00:00:00-04:00          94.940002       116.349998   \n",
       "2026-03-24 00:00:00-04:00          91.430000       116.529999   \n",
       "2026-03-25 00:00:00-04:00          89.970001       115.430000   \n",
       "2026-03-26 00:00:00-04:00          88.519997       115.059998   \n",
       "2026-03-27 00:00:00-04:00          86.169998       115.099998   \n",
       "2026-03-30 00:00:00-04:00          84.739998       114.980003   \n",
       "2026-03-31 00:00:00-04:00          83.000000       114.459999   \n",
       "2026-04-01 00:00:00-04:00          81.940002       114.209999   \n",
       "2026-04-02 00:00:00-04:00          80.379997       114.809998   \n",
       "2026-04-03 00:00:00-04:00          80.269997       115.389999   \n",
       "2026-04-06 00:00:00-04:00          79.019997       114.830002   \n",
       "2026-04-07 00:00:00-04:00          80.900002       116.309998   \n",
       "2026-04-08 00:00:00-04:00          79.139999       116.779999   \n",
       "2026-04-09 00:00:00-04:00          78.669998       117.029999   \n",
       "2026-04-10 00:00:00-04:00          78.650002       117.059998   \n",
       "2026-04-13 00:00:00-04:00          78.070000       117.790001   \n",
       "2026-04-14 00:00:00-04:00          77.400002       116.650002   \n",
       "2026-04-15 00:00:00-04:00          75.739998       116.559998   \n",
       "2026-04-16 00:00:00-04:00          75.209999       115.980003   \n",
       "2026-04-17 00:00:00-04:00          74.919998       115.320000   \n",
       "2026-04-20 00:00:00-04:00          73.400002       115.290001   \n",
       "2026-04-21 00:00:00-04:00          72.790001       114.139999   \n",
       "2026-04-22 00:00:00-04:00          72.199997       112.820000   \n",
       "2026-04-23 00:00:00-04:00          71.629997       111.230003   \n",
       "2026-04-24 00:00:00-04:00          70.559998       111.559998   \n",
       "\n",
       "                           Upper Bound (90%)  \n",
       "Date                                          \n",
       "2026-03-16 00:00:00-04:00         129.089996  \n",
       "2026-03-17 00:00:00-04:00         128.779999  \n",
       "2026-03-18 00:00:00-04:00         131.240005  \n",
       "2026-03-19 00:00:00-04:00         133.639999  \n",
       "2026-03-20 00:00:00-04:00         135.020004  \n",
       "2026-03-23 00:00:00-04:00         136.009995  \n",
       "2026-03-24 00:00:00-04:00         138.419998  \n",
       "2026-03-25 00:00:00-04:00         138.869995  \n",
       "2026-03-26 00:00:00-04:00         140.570007  \n",
       "2026-03-27 00:00:00-04:00         140.649994  \n",
       "2026-03-30 00:00:00-04:00         143.369995  \n",
       "2026-03-31 00:00:00-04:00         145.669998  \n",
       "2026-04-01 00:00:00-04:00         145.270004  \n",
       "2026-04-02 00:00:00-04:00         146.860001  \n",
       "2026-04-03 00:00:00-04:00         148.309998  \n",
       "2026-04-06 00:00:00-04:00         150.889999  \n",
       "2026-04-07 00:00:00-04:00         147.710007  \n",
       "2026-04-08 00:00:00-04:00         148.910004  \n",
       "2026-04-09 00:00:00-04:00         148.729996  \n",
       "2026-04-10 00:00:00-04:00         149.539993  \n",
       "2026-04-13 00:00:00-04:00         150.850006  \n",
       "2026-04-14 00:00:00-04:00         150.470001  \n",
       "2026-04-15 00:00:00-04:00         151.869995  \n",
       "2026-04-16 00:00:00-04:00         152.679993  \n",
       "2026-04-17 00:00:00-04:00         154.320007  \n",
       "2026-04-20 00:00:00-04:00         153.389999  \n",
       "2026-04-21 00:00:00-04:00         155.850006  \n",
       "2026-04-22 00:00:00-04:00         157.600006  \n",
       "2026-04-23 00:00:00-04:00         156.720001  \n",
       "2026-04-24 00:00:00-04:00         156.639999  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Display forecast data table\n",
    "forecast_table = pred_df[['timestamp', '0.1', 'predictions', '0.9']].copy()\n",
    "forecast_table.columns = ['Date', 'Lower Bound (10%)', 'Median Forecast', 'Upper Bound (90%)']\n",
    "forecast_table = forecast_table.set_index('Date')\n",
    "\n",
    "print(\"\\\\nDetailed 30-Day Forecast:\")\n",
    "forecast_table.round(2)\n",
    ""
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}