{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "22b1f6a0",
   "metadata": {},
   "source": [
    "# Quantitative Portfolio Research Pipeline\n",
    "\n",
    "This notebook shows the research code behind the portfolio project. The goal is not to claim a trading strategy. The goal is to show how a quantitative research idea can be turned into a repeatable, testable workflow.\n",
    "\n",
    "The workflow mirrors quant developer work:\n",
    "\n",
    "- Load market data into SQL.\n",
    "- Use Python, pandas, and NumPy to calculate returns and risk.\n",
    "- Use SciPy to solve constrained portfolio optimization problems.\n",
    "- Reproduce the minimum-variance optimization in MATLAB for cross-tool validation.\n",
    "- Run a walk-forward backtest so the model does not use future data.\n",
    "- Add tests for the calculations and assumptions.\n",
    "- Generate charts and a report that explain the results.\n",
    "\n",
    "Historical research only. No live trading or real-money execution is included."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f7a1acd",
   "metadata": {},
   "source": [
    "## Research Architecture\n",
    "\n",
    "```text\n",
    "Market data\n",
    "    |\n",
    "    v\n",
    "SQL data store\n",
    "    |\n",
    "    v\n",
    "Validation and cleaning\n",
    "    |\n",
    "    v\n",
    "Return and risk engine\n",
    "    |\n",
    "    v\n",
    "Constrained optimizer\n",
    "    |\n",
    "    v\n",
    "Walk-forward backtest\n",
    "    |\n",
    "    v\n",
    "Costs, benchmark comparison, and report\n",
    "```\n",
    "\n",
    "The key design choice is separation. Data ingestion, validation, calculations, optimization, backtesting, and reporting are separate pieces so the research can be changed and tested without rewriting the full project."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "9c876a11",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:06.786808Z",
     "iopub.status.busy": "2026-09-06T23:44:06.786501Z",
     "iopub.status.idle": "2026-09-06T23:44:08.232467Z",
     "shell.execute_reply": "2026-09-06T23:44:08.232121Z"
    }
   },
   "outputs": [],
   "source": [
    "from dataclasses import dataclass\n",
    "from pathlib import Path\n",
    "import sqlite3\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "try:\n",
    "    from scipy.optimize import minimize\n",
    "except ImportError as exc:\n",
    "    raise ImportError('Install scipy to run the optimization cells: pip install scipy') from exc\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.style.use('seaborn-v0_8-whitegrid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f5bb523e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.233951Z",
     "iopub.status.busy": "2026-09-06T23:44:08.233807Z",
     "iopub.status.idle": "2026-09-06T23:44:08.238020Z",
     "shell.execute_reply": "2026-09-06T23:44:08.237782Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ResearchConfig(tickers=('SPY', 'EWC', 'EFA', 'EEM', 'AGG', 'GLD'), benchmark='SPY', lookback_months=36, rebalance_frequency='Q', max_asset_weight=0.3, transaction_cost_bps=10.0, risk_free_rate=0.03, trading_days=252, database_path='portfolio_research.db')"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "@dataclass(frozen=True)\n",
    "class ResearchConfig:\n",
    "    tickers: tuple[str, ...] = ('SPY', 'EWC', 'EFA', 'EEM', 'AGG', 'GLD')\n",
    "    benchmark: str = 'SPY'\n",
    "    lookback_months: int = 36\n",
    "    rebalance_frequency: str = 'Q'\n",
    "    max_asset_weight: float = 0.30\n",
    "    transaction_cost_bps: float = 10.0\n",
    "    risk_free_rate: float = 0.03\n",
    "    trading_days: int = 252\n",
    "    database_path: str = 'portfolio_research.db'\n",
    "\n",
    "config = ResearchConfig()\n",
    "config"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e06cfcdd",
   "metadata": {},
   "source": [
    "## Market Data\n",
    "\n",
    "In the portfolio project, the report uses a frozen market snapshot so the results are reproducible. This notebook looks for a local CSV first. If it is not present, it creates a deterministic sample dataset so the notebook can still run during an interview without network access.\n",
    "\n",
    "A production version would replace the fallback with a scheduled data-freeze step from a licensed market data provider."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "348f6823",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.239155Z",
     "iopub.status.busy": "2026-09-06T23:44:08.239059Z",
     "iopub.status.idle": "2026-09-06T23:44:08.285815Z",
     "shell.execute_reply": "2026-09-06T23:44:08.285554Z"
    }
   },
   "outputs": [
    {
     "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>date</th>\n",
       "      <th>ticker</th>\n",
       "      <th>adj_close</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2018-01-01</td>\n",
       "      <td>AGG</td>\n",
       "      <td>99.630895</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018-01-02</td>\n",
       "      <td>AGG</td>\n",
       "      <td>99.071231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018-01-03</td>\n",
       "      <td>AGG</td>\n",
       "      <td>98.941628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018-01-04</td>\n",
       "      <td>AGG</td>\n",
       "      <td>99.240605</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018-01-05</td>\n",
       "      <td>AGG</td>\n",
       "      <td>99.564703</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        date ticker  adj_close\n",
       "0 2018-01-01    AGG  99.630895\n",
       "1 2018-01-02    AGG  99.071231\n",
       "2 2018-01-03    AGG  98.941628\n",
       "3 2018-01-04    AGG  99.240605\n",
       "4 2018-01-05    AGG  99.564703"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def make_demo_prices(config: ResearchConfig, start='2018-01-01', end='2025-12-31') -> pd.DataFrame:\n",
    "    rng = np.random.default_rng(42)\n",
    "    dates = pd.bdate_range(start, end)\n",
    "\n",
    "    annual_returns = {\n",
    "        'SPY': 0.105,\n",
    "        'EWC': 0.072,\n",
    "        'EFA': 0.066,\n",
    "        'EEM': 0.058,\n",
    "        'AGG': 0.028,\n",
    "        'GLD': 0.061,\n",
    "    }\n",
    "    annual_vols = {\n",
    "        'SPY': 0.175,\n",
    "        'EWC': 0.190,\n",
    "        'EFA': 0.185,\n",
    "        'EEM': 0.235,\n",
    "        'AGG': 0.060,\n",
    "        'GLD': 0.155,\n",
    "    }\n",
    "\n",
    "    # One shared equity factor, one defensive factor, and one commodity factor create realistic correlation patterns.\n",
    "    factors = rng.normal(size=(len(dates), 3))\n",
    "    rows = []\n",
    "    for ticker in config.tickers:\n",
    "        mu = annual_returns[ticker] / config.trading_days\n",
    "        sigma = annual_vols[ticker] / np.sqrt(config.trading_days)\n",
    "        if ticker in {'SPY', 'EWC', 'EFA', 'EEM'}:\n",
    "            shocks = 0.75 * factors[:, 0] + 0.15 * factors[:, 1] + 0.10 * rng.normal(size=len(dates))\n",
    "        elif ticker == 'AGG':\n",
    "            shocks = -0.10 * factors[:, 0] + 0.80 * factors[:, 1] + 0.10 * rng.normal(size=len(dates))\n",
    "        else:\n",
    "            shocks = 0.10 * factors[:, 0] + 0.20 * factors[:, 1] + 0.70 * factors[:, 2] + 0.10 * rng.normal(size=len(dates))\n",
    "\n",
    "        daily_returns = mu + sigma * shocks\n",
    "        prices = 100 * np.exp(np.cumsum(daily_returns))\n",
    "        rows.extend({'date': date, 'ticker': ticker, 'adj_close': price} for date, price in zip(dates, prices))\n",
    "\n",
    "    return pd.DataFrame(rows)\n",
    "\n",
    "def load_market_data(config: ResearchConfig) -> pd.DataFrame:\n",
    "    data_path = Path('data/etf_prices.csv')\n",
    "    if data_path.exists():\n",
    "        prices = pd.read_csv(data_path, parse_dates=['date'])\n",
    "    else:\n",
    "        prices = make_demo_prices(config)\n",
    "\n",
    "    prices = prices[['date', 'ticker', 'adj_close']].copy()\n",
    "    prices['date'] = pd.to_datetime(prices['date'])\n",
    "    prices = prices.sort_values(['ticker', 'date']).reset_index(drop=True)\n",
    "    return prices\n",
    "\n",
    "prices = load_market_data(config)\n",
    "prices.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e198c2f9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.287009Z",
     "iopub.status.busy": "2026-09-06T23:44:08.286901Z",
     "iopub.status.idle": "2026-09-06T23:44:08.295720Z",
     "shell.execute_reply": "2026-09-06T23:44:08.295480Z"
    }
   },
   "outputs": [
    {
     "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>check</th>\n",
       "      <th>passed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>required_columns</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>no_duplicate_ticker_date</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>positive_prices</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>all_expected_tickers_present</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>dates_increasing_by_ticker</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          check  passed\n",
       "0              required_columns    True\n",
       "1      no_duplicate_ticker_date    True\n",
       "2               positive_prices    True\n",
       "3  all_expected_tickers_present    True\n",
       "4    dates_increasing_by_ticker    True"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def validate_prices(prices: pd.DataFrame, expected_tickers: tuple[str, ...]) -> pd.DataFrame:\n",
    "    checks = []\n",
    "    checks.append(('required_columns', {'date', 'ticker', 'adj_close'}.issubset(prices.columns)))\n",
    "    checks.append(('no_duplicate_ticker_date', not prices.duplicated(['date', 'ticker']).any()))\n",
    "    checks.append(('positive_prices', bool((prices['adj_close'] > 0).all())))\n",
    "    checks.append(('all_expected_tickers_present', set(expected_tickers).issubset(set(prices['ticker']))))\n",
    "    checks.append(('dates_increasing_by_ticker', prices.groupby('ticker')['date'].apply(lambda s: s.is_monotonic_increasing).all()))\n",
    "\n",
    "    report = pd.DataFrame(checks, columns=['check', 'passed'])\n",
    "    if not report['passed'].all():\n",
    "        failed = report.loc[~report['passed'], 'check'].tolist()\n",
    "        raise ValueError(f'Data validation failed: {failed}')\n",
    "    return report\n",
    "\n",
    "validation_report = validate_prices(prices, config.tickers)\n",
    "validation_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "69cfcc6e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.296845Z",
     "iopub.status.busy": "2026-09-06T23:44:08.296748Z",
     "iopub.status.idle": "2026-09-06T23:44:08.348071Z",
     "shell.execute_reply": "2026-09-06T23:44:08.347819Z"
    }
   },
   "outputs": [
    {
     "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>rows_loaded</th>\n",
       "      <th>assets</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>12528</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   rows_loaded  assets\n",
       "0        12528       6"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def load_prices_to_sql(prices: pd.DataFrame, database_path: str) -> sqlite3.Connection:\n",
    "    con = sqlite3.connect(database_path)\n",
    "    con.executescript('''\n",
    "        DROP TABLE IF EXISTS prices;\n",
    "        CREATE TABLE prices (\n",
    "            date TEXT NOT NULL,\n",
    "            ticker TEXT NOT NULL,\n",
    "            adj_close REAL NOT NULL,\n",
    "            PRIMARY KEY (date, ticker)\n",
    "        );\n",
    "    ''')\n",
    "    prices.assign(date=prices['date'].dt.strftime('%Y-%m-%d')).to_sql('prices', con, if_exists='append', index=False)\n",
    "    return con\n",
    "\n",
    "con = load_prices_to_sql(prices, config.database_path)\n",
    "pd.read_sql_query('SELECT COUNT(*) AS rows_loaded, COUNT(DISTINCT ticker) AS assets FROM prices', con)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e2f47292",
   "metadata": {},
   "source": [
    "## SQL Examples\n",
    "\n",
    "These are the kinds of queries the project uses to make the data layer more than a CSV reader. They demonstrate joins, aggregations, common table expressions, and window functions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d74bd9b1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.349309Z",
     "iopub.status.busy": "2026-09-06T23:44:08.349201Z",
     "iopub.status.idle": "2026-09-06T23:44:08.375958Z",
     "shell.execute_reply": "2026-09-06T23:44:08.375728Z"
    }
   },
   "outputs": [
    {
     "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>date</th>\n",
       "      <th>ticker</th>\n",
       "      <th>daily_return</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2018-01-02</td>\n",
       "      <td>AGG</td>\n",
       "      <td>-0.005617</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018-01-02</td>\n",
       "      <td>EEM</td>\n",
       "      <td>0.007939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018-01-02</td>\n",
       "      <td>EFA</td>\n",
       "      <td>0.007096</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018-01-02</td>\n",
       "      <td>EWC</td>\n",
       "      <td>0.003716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018-01-02</td>\n",
       "      <td>GLD</td>\n",
       "      <td>-0.011755</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        date ticker  daily_return\n",
       "0 2018-01-02    AGG     -0.005617\n",
       "1 2018-01-02    EEM      0.007939\n",
       "2 2018-01-02    EFA      0.007096\n",
       "3 2018-01-02    EWC      0.003716\n",
       "4 2018-01-02    GLD     -0.011755"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "daily_return_query = '''\n",
    "WITH ordered_prices AS (\n",
    "    SELECT\n",
    "        date,\n",
    "        ticker,\n",
    "        adj_close,\n",
    "        LAG(adj_close) OVER (PARTITION BY ticker ORDER BY date) AS previous_close\n",
    "    FROM prices\n",
    "), daily_returns AS (\n",
    "    SELECT\n",
    "        date,\n",
    "        ticker,\n",
    "        adj_close / previous_close - 1.0 AS daily_return\n",
    "    FROM ordered_prices\n",
    "    WHERE previous_close IS NOT NULL\n",
    ")\n",
    "SELECT *\n",
    "FROM daily_returns\n",
    "ORDER BY date, ticker;\n",
    "'''\n",
    "\n",
    "returns_long = pd.read_sql_query(daily_return_query, con, parse_dates=['date'])\n",
    "returns_long.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "99270309",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.377072Z",
     "iopub.status.busy": "2026-09-06T23:44:08.376980Z",
     "iopub.status.idle": "2026-09-06T23:44:08.396308Z",
     "shell.execute_reply": "2026-09-06T23:44:08.396072Z"
    }
   },
   "outputs": [
    {
     "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>month</th>\n",
       "      <th>ticker</th>\n",
       "      <th>average_daily_return</th>\n",
       "      <th>observations</th>\n",
       "      <th>monthly_move_rank</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2018-01</td>\n",
       "      <td>SPY</td>\n",
       "      <td>0.001406</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018-01</td>\n",
       "      <td>GLD</td>\n",
       "      <td>0.001365</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018-01</td>\n",
       "      <td>EEM</td>\n",
       "      <td>0.001331</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018-02</td>\n",
       "      <td>GLD</td>\n",
       "      <td>-0.001059</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018-02</td>\n",
       "      <td>AGG</td>\n",
       "      <td>-0.000701</td>\n",
       "      <td>20</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2018-02</td>\n",
       "      <td>EEM</td>\n",
       "      <td>-0.000681</td>\n",
       "      <td>20</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2018-03</td>\n",
       "      <td>EEM</td>\n",
       "      <td>0.002169</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2018-03</td>\n",
       "      <td>SPY</td>\n",
       "      <td>0.001446</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2018-03</td>\n",
       "      <td>EWC</td>\n",
       "      <td>0.001131</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2018-04</td>\n",
       "      <td>EEM</td>\n",
       "      <td>-0.003224</td>\n",
       "      <td>21</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2018-04</td>\n",
       "      <td>SPY</td>\n",
       "      <td>-0.002167</td>\n",
       "      <td>21</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2018-04</td>\n",
       "      <td>EFA</td>\n",
       "      <td>-0.002141</td>\n",
       "      <td>21</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      month ticker  average_daily_return  observations  monthly_move_rank\n",
       "0   2018-01    SPY              0.001406            22                  1\n",
       "1   2018-01    GLD              0.001365            22                  2\n",
       "2   2018-01    EEM              0.001331            22                  3\n",
       "3   2018-02    GLD             -0.001059            20                  1\n",
       "4   2018-02    AGG             -0.000701            20                  2\n",
       "5   2018-02    EEM             -0.000681            20                  3\n",
       "6   2018-03    EEM              0.002169            22                  1\n",
       "7   2018-03    SPY              0.001446            22                  2\n",
       "8   2018-03    EWC              0.001131            22                  3\n",
       "9   2018-04    EEM             -0.003224            21                  1\n",
       "10  2018-04    SPY             -0.002167            21                  2\n",
       "11  2018-04    EFA             -0.002141            21                  3"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "monthly_volatility_query = '''\n",
    "WITH daily_returns AS (\n",
    "    SELECT\n",
    "        date,\n",
    "        ticker,\n",
    "        adj_close / LAG(adj_close) OVER (PARTITION BY ticker ORDER BY date) - 1.0 AS daily_return\n",
    "    FROM prices\n",
    "), monthly_stats AS (\n",
    "    SELECT\n",
    "        substr(date, 1, 7) AS month,\n",
    "        ticker,\n",
    "        AVG(daily_return) AS average_daily_return,\n",
    "        COUNT(*) AS observations\n",
    "    FROM daily_returns\n",
    "    WHERE daily_return IS NOT NULL\n",
    "    GROUP BY substr(date, 1, 7), ticker\n",
    "), ranked AS (\n",
    "    SELECT\n",
    "        month,\n",
    "        ticker,\n",
    "        average_daily_return,\n",
    "        observations,\n",
    "        RANK() OVER (PARTITION BY month ORDER BY ABS(average_daily_return) DESC) AS monthly_move_rank\n",
    "    FROM monthly_stats\n",
    ")\n",
    "SELECT *\n",
    "FROM ranked\n",
    "WHERE monthly_move_rank <= 3\n",
    "ORDER BY month, monthly_move_rank;\n",
    "'''\n",
    "\n",
    "pd.read_sql_query(monthly_volatility_query, con).head(12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "739c394a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.397410Z",
     "iopub.status.busy": "2026-09-06T23:44:08.397311Z",
     "iopub.status.idle": "2026-09-06T23:44:08.404551Z",
     "shell.execute_reply": "2026-09-06T23:44:08.404320Z"
    }
   },
   "outputs": [
    {
     "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>ticker</th>\n",
       "      <th>AGG</th>\n",
       "      <th>EEM</th>\n",
       "      <th>EFA</th>\n",
       "      <th>EWC</th>\n",
       "      <th>GLD</th>\n",
       "      <th>SPY</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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-01-02</th>\n",
       "      <td>-0.005617</td>\n",
       "      <td>0.007939</td>\n",
       "      <td>0.007096</td>\n",
       "      <td>0.003716</td>\n",
       "      <td>-0.011755</td>\n",
       "      <td>0.003709</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-03</th>\n",
       "      <td>-0.001308</td>\n",
       "      <td>0.003036</td>\n",
       "      <td>-0.001184</td>\n",
       "      <td>-0.000041</td>\n",
       "      <td>-0.000621</td>\n",
       "      <td>0.001424</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-04</th>\n",
       "      <td>0.003022</td>\n",
       "      <td>-0.006774</td>\n",
       "      <td>-0.007132</td>\n",
       "      <td>-0.008613</td>\n",
       "      <td>0.006950</td>\n",
       "      <td>-0.003927</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-05</th>\n",
       "      <td>0.003266</td>\n",
       "      <td>0.006021</td>\n",
       "      <td>0.001889</td>\n",
       "      <td>0.000288</td>\n",
       "      <td>0.006081</td>\n",
       "      <td>0.002743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-08</th>\n",
       "      <td>0.001439</td>\n",
       "      <td>-0.008319</td>\n",
       "      <td>-0.005695</td>\n",
       "      <td>-0.004603</td>\n",
       "      <td>-0.005709</td>\n",
       "      <td>-0.006282</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "ticker           AGG       EEM       EFA       EWC       GLD       SPY\n",
       "date                                                                  \n",
       "2018-01-02 -0.005617  0.007939  0.007096  0.003716 -0.011755  0.003709\n",
       "2018-01-03 -0.001308  0.003036 -0.001184 -0.000041 -0.000621  0.001424\n",
       "2018-01-04  0.003022 -0.006774 -0.007132 -0.008613  0.006950 -0.003927\n",
       "2018-01-05  0.003266  0.006021  0.001889  0.000288  0.006081  0.002743\n",
       "2018-01-08  0.001439 -0.008319 -0.005695 -0.004603 -0.005709 -0.006282"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "returns = returns_long.pivot(index='date', columns='ticker', values='daily_return').dropna()\n",
    "returns.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88377604",
   "metadata": {},
   "source": [
    "## Financial Calculations\n",
    "\n",
    "These functions convert raw returns into the financial inputs needed for the optimizer and report."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "88ac6ded",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.405574Z",
     "iopub.status.busy": "2026-09-06T23:44:08.405484Z",
     "iopub.status.idle": "2026-09-06T23:44:08.412853Z",
     "shell.execute_reply": "2026-09-06T23:44:08.412624Z"
    }
   },
   "outputs": [
    {
     "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>annual_return</th>\n",
       "      <th>annual_volatility</th>\n",
       "      <th>daily_var_95</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ticker</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AGG</th>\n",
       "      <td>0.015759</td>\n",
       "      <td>0.049364</td>\n",
       "      <td>0.005100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EEM</th>\n",
       "      <td>0.102031</td>\n",
       "      <td>0.178668</td>\n",
       "      <td>0.018259</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EFA</th>\n",
       "      <td>0.088668</td>\n",
       "      <td>0.140685</td>\n",
       "      <td>0.014190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EWC</th>\n",
       "      <td>0.092408</td>\n",
       "      <td>0.145111</td>\n",
       "      <td>0.014556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GLD</th>\n",
       "      <td>0.027424</td>\n",
       "      <td>0.117475</td>\n",
       "      <td>0.011973</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SPY</th>\n",
       "      <td>0.136535</td>\n",
       "      <td>0.132546</td>\n",
       "      <td>0.013124</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        annual_return  annual_volatility  daily_var_95\n",
       "ticker                                                \n",
       "AGG          0.015759           0.049364      0.005100\n",
       "EEM          0.102031           0.178668      0.018259\n",
       "EFA          0.088668           0.140685      0.014190\n",
       "EWC          0.092408           0.145111      0.014556\n",
       "GLD          0.027424           0.117475      0.011973\n",
       "SPY          0.136535           0.132546      0.013124"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def annualized_return(daily_returns: pd.Series, trading_days: int = 252) -> float:\n",
    "    compounded = (1 + daily_returns).prod()\n",
    "    years = len(daily_returns) / trading_days\n",
    "    return compounded ** (1 / years) - 1\n",
    "\n",
    "def annualized_volatility(daily_returns: pd.Series, trading_days: int = 252) -> float:\n",
    "    return daily_returns.std(ddof=1) * np.sqrt(trading_days)\n",
    "\n",
    "def sharpe_ratio(daily_returns: pd.Series, risk_free_rate: float, trading_days: int = 252) -> float:\n",
    "    excess_return = annualized_return(daily_returns, trading_days) - risk_free_rate\n",
    "    vol = annualized_volatility(daily_returns, trading_days)\n",
    "    return excess_return / vol\n",
    "\n",
    "def drawdown(daily_returns: pd.Series) -> pd.Series:\n",
    "    wealth = (1 + daily_returns).cumprod()\n",
    "    running_peak = wealth.cummax()\n",
    "    return wealth / running_peak - 1\n",
    "\n",
    "def value_at_risk(daily_returns: pd.Series, confidence: float = 0.95) -> float:\n",
    "    return -np.quantile(daily_returns, 1 - confidence)\n",
    "\n",
    "def portfolio_returns(returns: pd.DataFrame, weights: np.ndarray) -> pd.Series:\n",
    "    return pd.Series(returns.to_numpy() @ weights, index=returns.index)\n",
    "\n",
    "asset_summary = pd.DataFrame({\n",
    "    'annual_return': returns.apply(annualized_return),\n",
    "    'annual_volatility': returns.apply(annualized_volatility),\n",
    "    'daily_var_95': returns.apply(value_at_risk),\n",
    "})\n",
    "asset_summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3b6fd362",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.413907Z",
     "iopub.status.busy": "2026-09-06T23:44:08.413817Z",
     "iopub.status.idle": "2026-09-06T23:44:08.418668Z",
     "shell.execute_reply": "2026-09-06T23:44:08.418444Z"
    }
   },
   "outputs": [
    {
     "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>ticker</th>\n",
       "      <th>AGG</th>\n",
       "      <th>EEM</th>\n",
       "      <th>EFA</th>\n",
       "      <th>EWC</th>\n",
       "      <th>GLD</th>\n",
       "      <th>SPY</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ticker</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>AGG</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.095415</td>\n",
       "      <td>0.093975</td>\n",
       "      <td>0.094013</td>\n",
       "      <td>0.278163</td>\n",
       "      <td>0.095223</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EEM</th>\n",
       "      <td>0.095415</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.982585</td>\n",
       "      <td>0.982016</td>\n",
       "      <td>0.164117</td>\n",
       "      <td>0.983040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EFA</th>\n",
       "      <td>0.093975</td>\n",
       "      <td>0.982585</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.983307</td>\n",
       "      <td>0.163223</td>\n",
       "      <td>0.983809</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EWC</th>\n",
       "      <td>0.094013</td>\n",
       "      <td>0.982016</td>\n",
       "      <td>0.983307</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.163311</td>\n",
       "      <td>0.982480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GLD</th>\n",
       "      <td>0.278163</td>\n",
       "      <td>0.164117</td>\n",
       "      <td>0.163223</td>\n",
       "      <td>0.163311</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.159976</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SPY</th>\n",
       "      <td>0.095223</td>\n",
       "      <td>0.983040</td>\n",
       "      <td>0.983809</td>\n",
       "      <td>0.982480</td>\n",
       "      <td>0.159976</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "ticker       AGG       EEM       EFA       EWC       GLD       SPY\n",
       "ticker                                                            \n",
       "AGG     1.000000  0.095415  0.093975  0.094013  0.278163  0.095223\n",
       "EEM     0.095415  1.000000  0.982585  0.982016  0.164117  0.983040\n",
       "EFA     0.093975  0.982585  1.000000  0.983307  0.163223  0.983809\n",
       "EWC     0.094013  0.982016  0.983307  1.000000  0.163311  0.982480\n",
       "GLD     0.278163  0.164117  0.163223  0.163311  1.000000  0.159976\n",
       "SPY     0.095223  0.983040  0.983809  0.982480  0.159976  1.000000"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def portfolio_metrics(daily_returns: pd.Series, config: ResearchConfig) -> dict[str, float]:\n",
    "    return {\n",
    "        'annual_return': annualized_return(daily_returns, config.trading_days),\n",
    "        'annual_volatility': annualized_volatility(daily_returns, config.trading_days),\n",
    "        'sharpe_ratio': sharpe_ratio(daily_returns, config.risk_free_rate, config.trading_days),\n",
    "        'maximum_drawdown': drawdown(daily_returns).min(),\n",
    "        'daily_var_95': value_at_risk(daily_returns),\n",
    "    }\n",
    "\n",
    "covariance = returns.cov() * config.trading_days\n",
    "correlation = returns.corr()\n",
    "correlation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d35fa2e",
   "metadata": {},
   "source": [
    "## Optimization\n",
    "\n",
    "The optimizer solves two common portfolio problems:\n",
    "\n",
    "1. Minimum variance: find the lowest-risk mix.\n",
    "2. Maximum Sharpe: find the best return per unit of risk.\n",
    "\n",
    "Both are constrained: weights must add to 100%, no short positions are allowed, and no asset can be more than 30% of the portfolio."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "51576559",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.419803Z",
     "iopub.status.busy": "2026-09-06T23:44:08.419706Z",
     "iopub.status.idle": "2026-09-06T23:44:08.429854Z",
     "shell.execute_reply": "2026-09-06T23:44:08.429636Z"
    }
   },
   "outputs": [
    {
     "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>minimum_variance</th>\n",
       "      <th>maximum_sharpe</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AGG</th>\n",
       "      <td>3.000000e-01</td>\n",
       "      <td>0.005168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EEM</th>\n",
       "      <td>1.743972e-17</td>\n",
       "      <td>0.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EFA</th>\n",
       "      <td>1.082184e-01</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EWC</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.094832</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GLD</th>\n",
       "      <td>3.000000e-01</td>\n",
       "      <td>0.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SPY</th>\n",
       "      <td>2.917816e-01</td>\n",
       "      <td>0.300000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     minimum_variance  maximum_sharpe\n",
       "AGG      3.000000e-01        0.005168\n",
       "EEM      1.743972e-17        0.300000\n",
       "EFA      1.082184e-01        0.000000\n",
       "EWC      0.000000e+00        0.094832\n",
       "GLD      3.000000e-01        0.300000\n",
       "SPY      2.917816e-01        0.300000"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def optimize_portfolio(returns_window: pd.DataFrame, config: ResearchConfig, objective: str = 'max_sharpe') -> pd.Series:\n",
    "    assets = list(returns_window.columns)\n",
    "    n_assets = len(assets)\n",
    "    mean_returns = returns_window.mean().to_numpy() * config.trading_days\n",
    "    covariance = returns_window.cov().to_numpy() * config.trading_days\n",
    "\n",
    "    bounds = [(0.0, config.max_asset_weight) for _ in range(n_assets)]\n",
    "    constraints = [{'type': 'eq', 'fun': lambda w: np.sum(w) - 1.0}]\n",
    "    initial = np.repeat(1.0 / n_assets, n_assets)\n",
    "\n",
    "    def variance(weights: np.ndarray) -> float:\n",
    "        return float(weights.T @ covariance @ weights)\n",
    "\n",
    "    def negative_sharpe(weights: np.ndarray) -> float:\n",
    "        port_return = float(weights @ mean_returns)\n",
    "        port_volatility = np.sqrt(variance(weights))\n",
    "        return -(port_return - config.risk_free_rate) / port_volatility\n",
    "\n",
    "    objective_function = variance if objective == 'min_variance' else negative_sharpe\n",
    "    result = minimize(objective_function, initial, method='SLSQP', bounds=bounds, constraints=constraints)\n",
    "    if not result.success:\n",
    "        raise RuntimeError(result.message)\n",
    "\n",
    "    return pd.Series(result.x, index=assets).clip(lower=0)\n",
    "\n",
    "latest_window = returns.iloc[-config.trading_days * 3:]\n",
    "weights_min_var = optimize_portfolio(latest_window, config, objective='min_variance')\n",
    "weights_max_sharpe = optimize_portfolio(latest_window, config, objective='max_sharpe')\n",
    "\n",
    "pd.DataFrame({'minimum_variance': weights_min_var, 'maximum_sharpe': weights_max_sharpe})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "674dcb12",
   "metadata": {},
   "source": [
    "## MATLAB Comparison\n",
    "\n",
    "This is the same minimum-variance idea written in MATLAB style. The point is to show willingness to support quantitative models across the tools used by the team.\n",
    "\n",
    "```matlab\n",
    "% Minimum-variance portfolio with long-only weights and a 30% asset cap\n",
    "Sigma = cov(returns) * 252;\n",
    "n = size(Sigma, 1);\n",
    "\n",
    "f = zeros(n, 1);\n",
    "Aeq = ones(1, n);\n",
    "beq = 1;\n",
    "lb = zeros(n, 1);\n",
    "ub = 0.30 * ones(n, 1);\n",
    "\n",
    "options = optimoptions('quadprog', 'Display', 'off');\n",
    "weights = quadprog(2 * Sigma, f, [], [], Aeq, beq, lb, ub, [], options);\n",
    "```\n",
    "\n",
    "In the full project, the Python and MATLAB minimum-variance weights are compared within a tolerance so a model change can be reviewed instead of silently accepted."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91b7ef37",
   "metadata": {},
   "source": [
    "## Walk-Forward Backtest\n",
    "\n",
    "This is the most important part of the research design. The model estimates weights using only past data, then tests those weights on the next period. That prevents look-ahead bias."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "114dde76",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.430926Z",
     "iopub.status.busy": "2026-09-06T23:44:08.430831Z",
     "iopub.status.idle": "2026-09-06T23:44:08.466836Z",
     "shell.execute_reply": "2026-09-06T23:44:08.466592Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/bk/w06ly1yn09bgl47y13n5zpf00000gn/T/ipykernel_10877/2090468429.py:3: FutureWarning: 'Q' is deprecated and will be removed in a future version, please use 'QE' instead.\n",
      "  rebalance_dates = returns.resample(config.rebalance_frequency).last().index\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>optimized_portfolio</th>\n",
       "      <th>benchmark</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>annual_return</th>\n",
       "      <td>0.137317</td>\n",
       "      <td>0.212379</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>annual_volatility</th>\n",
       "      <td>0.110734</td>\n",
       "      <td>0.130285</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sharpe_ratio</th>\n",
       "      <td>0.969149</td>\n",
       "      <td>1.399848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>maximum_drawdown</th>\n",
       "      <td>-0.088619</td>\n",
       "      <td>-0.105223</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>daily_var_95</th>\n",
       "      <td>0.010812</td>\n",
       "      <td>0.012710</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   optimized_portfolio  benchmark\n",
       "annual_return                 0.137317   0.212379\n",
       "annual_volatility             0.110734   0.130285\n",
       "sharpe_ratio                  0.969149   1.399848\n",
       "maximum_drawdown             -0.088619  -0.105223\n",
       "daily_var_95                  0.010812   0.012710"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def walk_forward_backtest(returns: pd.DataFrame, config: ResearchConfig, objective: str = 'max_sharpe') -> tuple[pd.Series, pd.DataFrame]:\n",
    "    lookback_days = int(config.lookback_months * config.trading_days / 12)\n",
    "    rebalance_dates = returns.resample(config.rebalance_frequency).last().index\n",
    "    rebalance_dates = [date for date in rebalance_dates if date in returns.index and returns.index.get_loc(date) >= lookback_days]\n",
    "\n",
    "    portfolio_daily_returns = []\n",
    "    weight_records = []\n",
    "    previous_weights = pd.Series(0.0, index=returns.columns)\n",
    "    cost_rate = config.transaction_cost_bps / 10_000\n",
    "\n",
    "    for i, rebalance_date in enumerate(rebalance_dates[:-1]):\n",
    "        next_rebalance_date = rebalance_dates[i + 1]\n",
    "        end_position = returns.index.get_loc(rebalance_date)\n",
    "        train = returns.iloc[end_position - lookback_days:end_position + 1]\n",
    "        test = returns.loc[(returns.index > rebalance_date) & (returns.index <= next_rebalance_date)]\n",
    "\n",
    "        weights = optimize_portfolio(train, config, objective=objective)\n",
    "        turnover = (weights - previous_weights).abs().sum()\n",
    "        transaction_cost = turnover * cost_rate\n",
    "\n",
    "        period_returns = portfolio_returns(test, weights.to_numpy())\n",
    "        if not period_returns.empty:\n",
    "            period_returns.iloc[0] -= transaction_cost\n",
    "            portfolio_daily_returns.append(period_returns)\n",
    "\n",
    "        weight_records.append({\n",
    "            'rebalance_date': rebalance_date,\n",
    "            'next_rebalance_date': next_rebalance_date,\n",
    "            'turnover': turnover,\n",
    "            'transaction_cost': transaction_cost,\n",
    "            **weights.to_dict(),\n",
    "        })\n",
    "        previous_weights = weights\n",
    "\n",
    "    return pd.concat(portfolio_daily_returns).sort_index(), pd.DataFrame(weight_records)\n",
    "\n",
    "strategy_returns, rebalance_history = walk_forward_backtest(returns, config)\n",
    "benchmark_returns = returns.loc[strategy_returns.index, config.benchmark]\n",
    "\n",
    "pd.DataFrame({\n",
    "    'optimized_portfolio': portfolio_metrics(strategy_returns, config),\n",
    "    'benchmark': portfolio_metrics(benchmark_returns, config),\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6f8db09e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.468013Z",
     "iopub.status.busy": "2026-09-06T23:44:08.467908Z",
     "iopub.status.idle": "2026-09-06T23:44:08.472629Z",
     "shell.execute_reply": "2026-09-06T23:44:08.472403Z"
    }
   },
   "outputs": [
    {
     "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>rebalance_date</th>\n",
       "      <th>next_rebalance_date</th>\n",
       "      <th>turnover</th>\n",
       "      <th>transaction_cost</th>\n",
       "      <th>AGG</th>\n",
       "      <th>EEM</th>\n",
       "      <th>EFA</th>\n",
       "      <th>EWC</th>\n",
       "      <th>GLD</th>\n",
       "      <th>SPY</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2024-09-30</td>\n",
       "      <td>2024-12-31</td>\n",
       "      <td>1.200000e+00</td>\n",
       "      <td>1.200000e-03</td>\n",
       "      <td>1.892070e-18</td>\n",
       "      <td>0.1</td>\n",
       "      <td>3.000000e-01</td>\n",
       "      <td>0.3</td>\n",
       "      <td>2.815601e-17</td>\n",
       "      <td>0.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2024-12-31</td>\n",
       "      <td>2025-03-31</td>\n",
       "      <td>4.000000e-01</td>\n",
       "      <td>4.000000e-04</td>\n",
       "      <td>9.385261e-17</td>\n",
       "      <td>0.3</td>\n",
       "      <td>1.000000e-01</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2025-03-31</td>\n",
       "      <td>2025-06-30</td>\n",
       "      <td>4.902974e-16</td>\n",
       "      <td>4.902974e-19</td>\n",
       "      <td>3.376668e-17</td>\n",
       "      <td>0.3</td>\n",
       "      <td>1.000000e-01</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2025-06-30</td>\n",
       "      <td>2025-09-30</td>\n",
       "      <td>7.311761e-17</td>\n",
       "      <td>7.311761e-20</td>\n",
       "      <td>3.003800e-17</td>\n",
       "      <td>0.3</td>\n",
       "      <td>1.000000e-01</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2025-09-30</td>\n",
       "      <td>2025-12-31</td>\n",
       "      <td>2.000000e-01</td>\n",
       "      <td>2.000000e-04</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.3</td>\n",
       "      <td>3.469447e-17</td>\n",
       "      <td>0.3</td>\n",
       "      <td>1.000000e-01</td>\n",
       "      <td>0.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   rebalance_date next_rebalance_date      turnover  transaction_cost  \\\n",
       "10     2024-09-30          2024-12-31  1.200000e+00      1.200000e-03   \n",
       "11     2024-12-31          2025-03-31  4.000000e-01      4.000000e-04   \n",
       "12     2025-03-31          2025-06-30  4.902974e-16      4.902974e-19   \n",
       "13     2025-06-30          2025-09-30  7.311761e-17      7.311761e-20   \n",
       "14     2025-09-30          2025-12-31  2.000000e-01      2.000000e-04   \n",
       "\n",
       "             AGG  EEM           EFA  EWC           GLD  SPY  \n",
       "10  1.892070e-18  0.1  3.000000e-01  0.3  2.815601e-17  0.3  \n",
       "11  9.385261e-17  0.3  1.000000e-01  0.3  0.000000e+00  0.3  \n",
       "12  3.376668e-17  0.3  1.000000e-01  0.3  0.000000e+00  0.3  \n",
       "13  3.003800e-17  0.3  1.000000e-01  0.3  0.000000e+00  0.3  \n",
       "14  0.000000e+00  0.3  3.469447e-17  0.3  1.000000e-01  0.3  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rebalance_history.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "93e17430",
   "metadata": {},
   "source": [
    "## MATLAB Visualization Equivalents\n",
    "\n",
    "The rendered charts below are produced from the same research data in the notebook. Since the role specifically mentions MATLAB, this section shows how the visual review can be reproduced in MATLAB once the return series, benchmark series, correlation matrix, and optimized weights are exported from the Python pipeline.\n",
    "\n",
    "```matlab\n",
    "% Growth of $1\n",
    "portfolioGrowth = cumprod(1 + strategyReturns);\n",
    "benchmarkGrowth = cumprod(1 + benchmarkReturns);\n",
    "plot(dates, portfolioGrowth, LineWidth, 2); hold on;\n",
    "plot(dates, benchmarkGrowth, LineWidth, 2);\n",
    "title(Growth of $1);\n",
    "ylabel(Portfolio value);\n",
    "legend(Optimized portfolio, Benchmark);\n",
    "\n",
    "% Drawdown\n",
    "portfolioDrawdown = portfolioGrowth ./ cummax(portfolioGrowth) - 1;\n",
    "benchmarkDrawdown = benchmarkGrowth ./ cummax(benchmarkGrowth) - 1;\n",
    "figure;\n",
    "plot(dates, portfolioDrawdown, LineWidth, 2); hold on;\n",
    "plot(dates, benchmarkDrawdown, LineWidth, 2);\n",
    "title(Drawdown);\n",
    "ylabel(Drawdown);\n",
    "legend(Optimized portfolio, Benchmark);\n",
    "\n",
    "% Correlation heatmap\n",
    "figure;\n",
    "heatmap(assetNames, assetNames, correlationMatrix);\n",
    "title(Daily Return Correlation);\n",
    "\n",
    "% Latest optimized weights\n",
    "figure;\n",
    "barh(latestWeights);\n",
    "yticklabels(assetNames);\n",
    "xlabel(Portfolio weight);\n",
    "title(Latest Optimized Weights);\n",
    "```\n",
    "\n",
    "This helps show that the research is not locked to one plotting library. Python handles the pipeline and MATLAB can reproduce the model checks and charting workflow used by a quant team.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ee97643d",
   "metadata": {},
   "source": [
    "## Visual Checks\n",
    "\n",
    "These outputs make the research easier to review. The charts show whether the optimized portfolio grows differently from the benchmark, how painful the worst declines were, how the assets move together, and whether the latest allocation is concentrated.\n",
    "\n",
    "For the interview, this is the visualization story: the same calculations feed both the Python report and the MATLAB-style review section above. That shows the model can be inspected across tools, not just hidden behind one charting package.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "97c54b64",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.473678Z",
     "iopub.status.busy": "2026-09-06T23:44:08.473589Z",
     "iopub.status.idle": "2026-09-06T23:44:08.617903Z",
     "shell.execute_reply": "2026-09-06T23:44:08.617654Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1100x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "growth = pd.DataFrame({\n",
    "    'Optimized portfolio': (1 + strategy_returns).cumprod(),\n",
    "    f'Benchmark ({config.benchmark})': (1 + benchmark_returns).cumprod(),\n",
    "})\n",
    "\n",
    "ax = growth.plot(figsize=(11, 5), linewidth=2)\n",
    "ax.set_title('Growth of $1')\n",
    "ax.set_ylabel('Portfolio value')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "1f54e734",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.619147Z",
     "iopub.status.busy": "2026-09-06T23:44:08.619042Z",
     "iopub.status.idle": "2026-09-06T23:44:08.686430Z",
     "shell.execute_reply": "2026-09-06T23:44:08.686144Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1100x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dd = pd.DataFrame({\n",
    "    'Optimized portfolio': drawdown(strategy_returns),\n",
    "    f'Benchmark ({config.benchmark})': drawdown(benchmark_returns),\n",
    "})\n",
    "\n",
    "ax = dd.plot(figsize=(11, 5), linewidth=2)\n",
    "ax.set_title('Drawdown')\n",
    "ax.set_ylabel('Drawdown')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "1efd535c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.687690Z",
     "iopub.status.busy": "2026-09-06T23:44:08.687589Z",
     "iopub.status.idle": "2026-09-06T23:44:08.756489Z",
     "shell.execute_reply": "2026-09-06T23:44:08.756203Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(7, 6))\n",
    "im = ax.imshow(correlation, cmap='Greens', vmin=-1, vmax=1)\n",
    "ax.set_xticks(range(len(correlation.columns)), correlation.columns, rotation=45)\n",
    "ax.set_yticks(range(len(correlation.index)), correlation.index)\n",
    "for i in range(len(correlation.index)):\n",
    "    for j in range(len(correlation.columns)):\n",
    "        ax.text(j, i, f'{correlation.iloc[i, j]:.2f}', ha='center', va='center', fontsize=9)\n",
    "fig.colorbar(im, ax=ax, fraction=0.046, pad=0.04)\n",
    "ax.set_title('Daily Return Correlation')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e6c743e7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.757659Z",
     "iopub.status.busy": "2026-09-06T23:44:08.757577Z",
     "iopub.status.idle": "2026-09-06T23:44:08.792279Z",
     "shell.execute_reply": "2026-09-06T23:44:08.792006Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "latest_weights = rebalance_history.set_index('rebalance_date')[list(config.tickers)].iloc[-1].sort_values(ascending=True)\n",
    "ax = latest_weights.plot(kind='barh', figsize=(8, 4), color='#176b55')\n",
    "ax.set_title('Latest Optimized Weights')\n",
    "ax.set_xlabel('Portfolio weight')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81c5a4fa",
   "metadata": {},
   "source": [
    "## How To Read The Visuals\n",
    "\n",
    "- Growth of $1 shows whether the optimized strategy added value versus the benchmark over time.\n",
    "- Drawdown shows the worst loss from a previous high, which is often more intuitive than volatility alone.\n",
    "- Correlation shows diversification quality. Lower correlation means the assets are less likely to fall together.\n",
    "- Optimized weights show the final portfolio recommendation under the 30% maximum asset constraint.\n",
    "\n",
    "A quant developer is not only expected to calculate these metrics. They also need to make the results reviewable by researchers, portfolio managers, and risk teams.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f34bd58",
   "metadata": {},
   "source": [
    "## Calculation Tests\n",
    "\n",
    "These are small examples of the tests that matter in a quant project. The goal is to make sure the simple financial building blocks are correct before trusting the larger optimization and backtest."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "40a4708d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-09-06T23:44:08.793445Z",
     "iopub.status.busy": "2026-09-06T23:44:08.793361Z",
     "iopub.status.idle": "2026-09-06T23:44:08.799336Z",
     "shell.execute_reply": "2026-09-06T23:44:08.799081Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'All notebook checks passed'"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def test_daily_return_math():\n",
    "    simple_prices = pd.Series([100, 110])\n",
    "    calculated = simple_prices.pct_change().iloc[1]\n",
    "    assert np.isclose(calculated, 0.10)\n",
    "\n",
    "def test_portfolio_return_math():\n",
    "    simple_returns = pd.DataFrame({'Asset A': [0.10], 'Asset B': [0.00]})\n",
    "    weights = np.array([0.50, 0.50])\n",
    "    calculated = portfolio_returns(simple_returns, weights).iloc[0]\n",
    "    assert np.isclose(calculated, 0.05)\n",
    "\n",
    "def test_optimizer_constraints():\n",
    "    weights = optimize_portfolio(latest_window, config, objective='max_sharpe')\n",
    "    assert np.isclose(weights.sum(), 1.0, atol=1e-6)\n",
    "    assert (weights >= -1e-8).all()\n",
    "    assert (weights <= config.max_asset_weight + 1e-6).all()\n",
    "\n",
    "def test_walk_forward_chronology():\n",
    "    assert (rebalance_history['next_rebalance_date'] > rebalance_history['rebalance_date']).all()\n",
    "\n",
    "for test in [\n",
    "    test_daily_return_math,\n",
    "    test_portfolio_return_math,\n",
    "    test_optimizer_constraints,\n",
    "    test_walk_forward_chronology,\n",
    "]:\n",
    "    test()\n",
    "\n",
    "'All notebook checks passed'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "188ed10d",
   "metadata": {},
   "source": [
    "## What The Project Demonstrates\n",
    "\n",
    "This project demonstrates the practical parts of quant development:\n",
    "\n",
    "- Translating an investment question into calculations and code.\n",
    "- Using SQL as a structured data layer, not just relying on spreadsheets.\n",
    "- Using Python and pandas for financial analytics.\n",
    "- Using optimization under realistic constraints.\n",
    "- Avoiding look-ahead bias with walk-forward testing.\n",
    "- Accounting for transaction costs and turnover.\n",
    "- Reproducing a model in MATLAB for validation.\n",
    "- Writing tests around financial calculations and model assumptions.\n",
    "\n",
    "The result is a small research pipeline that could be explained, rerun, reviewed, and modified by a quantitative research team."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.5"
  },
  "title": "Quantitative Portfolio Research Pipeline"
 },
 "nbformat": 4,
 "nbformat_minor": 5
}