{ "cells": [ { "cell_type": "markdown", "id": "ba32eb4a", "metadata": {}, "source": [ "# Backtest and Performance\n", "\n", "## Purpose\n", "\n", "This is the main portfolio notebook. We take the sector-neutralized momentum signal from notebook 03, form portfolio weights, subtract transaction costs, and measure the result.\n", "\n", "A portfolio is a weight vector. If $w_t$ is the portfolio chosen at rebalance date $t$, and $r_{t+1}$ is the next month's return vector, then the portfolio return is\n", "\n", "$$r_{p,t+1}=w_t^\\top r_{t+1}.$$\n", "\n", "The goals are:\n", "1. Build top-decile long-only and long-short portfolios from the momentum signal.\n", "2. Track one-way turnover explicitly: $\\tfrac{1}{2}\\|w_t-w_{t-1}\\|_1$.\n", "3. Subtract transaction costs proportional to turnover.\n", "4. Compute Sharpe, Sortino, max drawdown, Calmar, and active return versus the equal-weight universe.\n", "5. Run walk-forward checks across 5-year windows.\n", "6. Estimate Fama-French alpha with an OLS regression.\n", "7. Stress-test survivorship bias by asking how much missing-name drag would erase the alpha.\n", "\n", "### Terms used in this notebook\n", "\n", "| Term | Meaning |\n", "|------|---------|\n", "| **Long** | Holding a stock with positive weight $w_i>0$ |\n", "| **Short** | Holding a negative weight $w_i<0$ |\n", "| **Long-only portfolio** | Weight vector with $w_i\\ge 0$ and $\\sum_i w_i=1$ |\n", "| **Long-short portfolio** | Long winners and short losers; roughly dollar-neutral with $\\sum_i w_i=0$ |\n", "| **Decile** | Top or bottom 10% of stocks by signal rank |\n", "| **Basis point (bp)** | 1 bp = 0.01%; 5 bps = 0.05% |\n", "| **Turnover** | One-way turnover $\\tfrac{1}{2}\\|w_t-w_{t-1}\\|_1$ |\n", "| **Transaction cost** | Cost rate times one-way turnover |\n", "| **Sharpe ratio** | Annualized mean return divided by annualized volatility |\n", "| **Sortino ratio** | Similar to Sharpe, but only downside moves enter the denominator |\n", "| **Max drawdown** | Worst percentage decline from a previous wealth peak |\n", "| **Alpha** | Regression intercept after controlling for benchmark factors |\n", "| **Beta** | Regression loading on a benchmark factor |\n", "| **Fama-French factors** | Standard market, size, value, and momentum benchmark returns |\n", "| **Active return / IR** | Portfolio return minus benchmark return; IR = active return / tracking error |\n", "| **Survivorship bias** ↻ | Tested here with synthetic return drag |\n", "\n", "## Outputs\n", "\n", "Equity curves, drawdowns, performance tables, Fama-French regressions, survivorship sensitivity tables, and `backtest_returns.csv` for the risk notebook.\n", "\n", "## Notebook Structure\n", "1. [Setup and Imports](#setup-and-imports)\n", "2. [Load Data and Benchmark Factors](#load-data-and-benchmark-factors)\n", "3. [Portfolio Formation](#portfolio-formation)\n", "4. [Turnover and Transaction Costs](#turnover-and-transaction-costs)\n", "5. [Performance Metrics](#performance-metrics)\n", "6. [Walk-Forward Analysis](#walk-forward-analysis)\n", "7. [Fama–French Alpha](#famafrench-alpha)\n", "8. [Survivorship Bias Sensitivity](#survivorship-bias-sensitivity)\n", "9. [Conclusion](#conclusion)" ] }, { "cell_type": "code", "execution_count": 1, "id": "b2db11de", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:19.758791Z", "iopub.status.busy": "2026-07-31T12:20:19.757884Z", "iopub.status.idle": "2026-07-31T12:20:21.098020Z", "shell.execute_reply": "2026-07-31T12:20:21.097429Z" } }, "outputs": [], "source": [ "\"\"\"\n", "==================================\n", "Setup and imports\n", "==================================\n", "\"\"\"\n", "import pandas as pd\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "import os\n", "\n", "pd.set_option('display.max_columns', None)\n", "pd.set_option('display.max_rows', 100)\n", "\n", "palette = ['steelblue', 'coral', 'seagreen']\n", "\n", "os.makedirs('../data/processed', exist_ok=True)\n", "os.makedirs('../images/04_backtest', exist_ok=True)\n", "\n", "RANDOM_STATE = 3\n", "TRANSACTION_COST_BPS = 5 # 5 bps per unit of one-way turnover\n", "REBALANCE_FREQ = 'ME' # Monthly" ] }, { "cell_type": "markdown", "id": "293597e5", "metadata": {}, "source": [ "## Load Data and Benchmark Factors" ] }, { "cell_type": "code", "execution_count": 2, "id": "9ffbee22", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:21.100126Z", "iopub.status.busy": "2026-07-31T12:20:21.099825Z", "iopub.status.idle": "2026-07-31T12:20:21.176089Z", "shell.execute_reply": "2026-07-31T12:20:21.175603Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Momentum: (251, 501)\n", "Returns: (251, 501)\n" ] } ], "source": [ "\"\"\"\n", "==================================\n", "Load momentum signal and returns\n", "==================================\n", "\"\"\"\n", "df_momentum = pd.read_csv('../data/processed/momentum_signal.csv', index_col=0, parse_dates=True)\n", "df_returns = pd.read_csv('../data/processed/returns_monthly.csv', index_col=0, parse_dates=True)\n", "\n", "print(f\"Momentum: {df_momentum.shape}\")\n", "print(f\"Returns: {df_returns.shape}\")" ] }, { "cell_type": "markdown", "id": "05c2fc00", "metadata": {}, "source": [ "## Benchmark Factors and Alpha\n", "\n", "A 20% portfolio return sounds good, but it may not be stock-picking skill. If the market returned 18% and the portfolio had high market beta, most of the return may be ordinary market exposure.\n", "\n", "To separate those effects, we use the Fama-French benchmark factors. The regression later in this notebook is:\n", "\n", "$$r_p - r_f = \\alpha + \\beta_1\\mathrm{MKT} + \\beta_2\\mathrm{SMB} + \\beta_3\\mathrm{HML} + \\beta_4\\mathrm{MOM} + \\varepsilon.$$\n", "\n", "| Factor | Symbol | What it captures |\n", "|--------|--------|-----------------|\n", "| **Market** | MKT-RF | Market excess return above the risk-free rate |\n", "| **Size** | SMB | Small-minus-big stock return spread |\n", "| **Value** | HML | High-minus-low book-to-market return spread |\n", "| **Momentum** | MOM | Winner-minus-loser momentum factor |\n", "| **Risk-free rate** | RF | Monthly T-bill rate used to compute excess returns |\n", "\n", "The betas measure exposure to known return drivers. The alpha is the intercept: the average monthly return left over after those exposures are accounted for. A positive alpha with a large t-statistic is evidence that the strategy is doing more than taking standard factor risk.\n", "\n", "These factors are loaded from the Kenneth French Data Library, using the local cache when available." ] }, { "cell_type": "code", "execution_count": 3, "id": "5bfbd826", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:21.177834Z", "iopub.status.busy": "2026-07-31T12:20:21.177665Z", "iopub.status.idle": "2026-07-31T12:20:21.192027Z", "shell.execute_reply": "2026-07-31T12:20:21.191419Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "FF factors: (257, 5)\n", "Columns: ['Mkt-RF', 'SMB', 'HML', 'RF', 'Mom']\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Mkt-RFSMBHMLRFMom
Date
2005-01-01-0.0275-0.01660.02060.00160.0312
2005-02-010.0188-0.00570.01410.00160.0343
2005-03-01-0.0194-0.01410.02070.00210.0043
2005-04-01-0.0261-0.03930.00050.0021-0.0070
2005-05-010.03650.0286-0.00580.00240.0037
\n", "
" ], "text/plain": [ " Mkt-RF SMB HML RF Mom\n", "Date \n", "2005-01-01 -0.0275 -0.0166 0.0206 0.0016 0.0312\n", "2005-02-01 0.0188 -0.0057 0.0141 0.0016 0.0343\n", "2005-03-01 -0.0194 -0.0141 0.0207 0.0021 0.0043\n", "2005-04-01 -0.0261 -0.0393 0.0005 0.0021 -0.0070\n", "2005-05-01 0.0365 0.0286 -0.0058 0.0024 0.0037" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "\"\"\"\n", "==================================\n", "Download Fama-French 3-factor + Momentum from Ken French data library\n", "==================================\n", "\"\"\"\n", "ff_path = '../data/raw/ff_factors.csv'\n", "\n", "if os.path.exists(ff_path):\n", " df_ff = pd.read_csv(ff_path, index_col=0, parse_dates=True)\n", "else:\n", " import pandas_datareader as pdr\n", "\n", " print(\"Downloading Fama-French factors...\")\n", " # 3-factor monthly\n", " df_ff3 = pdr.famafrench.FamaFrenchReader('F-F_Research_Data_Factors', start='2005-01-01').read()[0]\n", " df_ff3.index = df_ff3.index.to_timestamp()\n", " df_ff3 = df_ff3 / 100 # Convert from percent to decimal\n", " \n", " # Momentum monthly\n", " df_mom = pdr.famafrench.FamaFrenchReader('F-F_Momentum_Factor', start='2005-01-01').read()[0]\n", " df_mom.index = df_mom.index.to_timestamp()\n", " df_mom = df_mom / 100\n", " \n", " df_ff = df_ff3.join(df_mom)\n", " df_ff.to_csv(ff_path)\n", " print(f\"Saved to {ff_path}\")\n", "\n", "print(f\"FF factors: {df_ff.shape}\")\n", "print(f\"Columns: {list(df_ff.columns)}\")\n", "df_ff.head()" ] }, { "cell_type": "markdown", "id": "fcbaf1c1", "metadata": {}, "source": [ "## Portfolio Formation\n", "\n", "At each rebalance date, we rank stocks by the momentum signal (the vector $c_t$) and form two portfolios:\n", "- **Long-only**: equal-weight the top decile (top 10% of stocks by momentum score). This is a sparse weight vector with $w_i = 1/k$ for the top $k$ stocks and $w_i = 0$ for the rest.\n", "- **Long-short (comparison)**: long the top decile, short the bottom decile, equal-weighted on each side with $\\sum w_i = 0$ (dollar-neutral).\n", "\n" ] }, { "cell_type": "code", "execution_count": 4, "id": "8691aa8a", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:21.193875Z", "iopub.status.busy": "2026-07-31T12:20:21.193696Z", "iopub.status.idle": "2026-07-31T12:20:21.438807Z", "shell.execute_reply": "2026-07-31T12:20:21.438161Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Backtest period: 2006-02-28 00:00:00 to 2025-12-31 00:00:00\n", "Number of rebalances: 239\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
longshortls
count239.0000239.0000239.0000
mean0.01640.0168-0.0004
std0.05600.07000.0479
min-0.1796-0.2209-0.4085
25%-0.0116-0.0211-0.0214
50%0.01760.01420.0023
75%0.04970.04860.0250
max0.17270.47960.1008
\n", "
" ], "text/plain": [ " long short ls\n", "count 239.0000 239.0000 239.0000\n", "mean 0.0164 0.0168 -0.0004\n", "std 0.0560 0.0700 0.0479\n", "min -0.1796 -0.2209 -0.4085\n", "25% -0.0116 -0.0211 -0.0214\n", "50% 0.0176 0.0142 0.0023\n", "75% 0.0497 0.0486 0.0250\n", "max 0.1727 0.4796 0.1008" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "\"\"\"\n", "==================================\n", "Form top-decile long and long-short portfolios\n", "==================================\n", "\n", "At each rebalance date:\n", " - rank stocks by momentum score\n", " - long the top decile, equal-weighted\n", " - short the bottom decile, equal-weighted (for L/S portfolio)\n", "\n", "We trade on t+1 to avoid look-ahead bias: signal computed at month-end t,\n", "returns realized over month t+1.\n", "\"\"\"\n", "def form_decile_portfolios(signal_df, return_df, decile=0.1):\n", " \"\"\"Form long (top decile) and short (bottom decile) portfolios.\"\"\"\n", " common_dates = signal_df.index.intersection(return_df.index)\n", " common_tickers = signal_df.columns.intersection(return_df.columns)\n", " \n", " signal_df = signal_df.loc[common_dates, common_tickers]\n", " return_df = return_df.loc[common_dates, common_tickers]\n", " \n", " long_returns = []\n", " short_returns = []\n", " ls_returns = []\n", " long_holdings = []\n", " short_holdings = []\n", " rebalance_dates = []\n", " \n", " for i in range(len(common_dates) - 1):\n", " date = common_dates[i]\n", " next_date = common_dates[i + 1]\n", " \n", " scores = signal_df.loc[date].dropna()\n", " if len(scores) < 50:\n", " continue\n", " \n", " n_long = max(int(len(scores) * decile), 1)\n", " n_short = max(int(len(scores) * decile), 1)\n", " \n", " ranked = scores.sort_values(ascending=False)\n", " long_tickers = ranked.head(n_long).index.tolist()\n", " short_tickers = ranked.tail(n_short).index.tolist()\n", " \n", " # Returns realized over next month\n", " next_rets = return_df.loc[next_date]\n", " \n", " long_ret = next_rets[long_tickers].mean()\n", " short_ret = next_rets[short_tickers].mean()\n", " ls_ret = long_ret - short_ret\n", " \n", " long_returns.append(long_ret)\n", " short_returns.append(short_ret)\n", " ls_returns.append(ls_ret)\n", " long_holdings.append(long_tickers)\n", " short_holdings.append(short_tickers)\n", " rebalance_dates.append(next_date)\n", " \n", " df_port = pd.DataFrame({\n", " 'long': long_returns,\n", " 'short': short_returns,\n", " 'ls': ls_returns,\n", " 'long_holdings': long_holdings,\n", " 'short_holdings': short_holdings\n", " }, index=pd.DatetimeIndex(rebalance_dates))\n", " \n", " return df_port\n", "\n", "df_port = form_decile_portfolios(df_momentum, df_returns)\n", "print(f\"Backtest period: {df_port.index.min()} to {df_port.index.max()}\")\n", "print(f\"Number of rebalances: {len(df_port)}\")\n", "df_port[['long', 'short', 'ls']].describe().round(4)" ] }, { "cell_type": "markdown", "id": "a438dcfb", "metadata": {}, "source": [ "## Turnover and Transaction Costs\n", "\n", "One-way turnover measures how much of the portfolio has to be traded at each rebalance:\n", "\n", "$$\\text{turnover}_t=\\frac{1}{2}\\|w_t-w_{t-1}\\|_1.$$\n", "\n", "For a long-only portfolio, zero means nothing changed. A value near 1 means the portfolio was almost completely replaced. For the long-short book, we compute one-way turnover on the long side and short side and add them.\n", "\n", "Transaction costs are proportional to turnover:\n", "\n", "$$\\text{cost}_t=c\\times\\text{turnover}_t,$$\n", "\n", "where $c=5$ bps, or $0.0005$, for liquid US large-cap names. Net return is gross return minus cost. The first rebalance includes the initial cost of buying the portfolio." ] }, { "cell_type": "code", "execution_count": 5, "id": "7f3b7fe9", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:21.440964Z", "iopub.status.busy": "2026-07-31T12:20:21.440773Z", "iopub.status.idle": "2026-07-31T12:20:22.027373Z", "shell.execute_reply": "2026-07-31T12:20:22.026659Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Average long turnover (one-way): 0.289\n", "Average L/S turnover (one-way): 0.596\n" ] } ], "source": [ "\"\"\"\n", "==================================\n", "Compute one-way turnover at each rebalance\n", "==================================\n", "\"\"\"\n", "def _equal_weight_vector(holdings):\n", " \"\"\"Equal-weight vector for a list of holdings.\"\"\"\n", " if len(holdings) == 0:\n", " return pd.Series(dtype=float)\n", " return pd.Series(1.0 / len(holdings), index=pd.Index(holdings), dtype=float)\n", "\n", "def compute_turnover(holdings_series):\n", " \"\"\"Compute one-way turnover from actual target weight vectors.\"\"\"\n", " turnovers = []\n", " prev_w = None\n", " for holdings in holdings_series:\n", " curr_w = _equal_weight_vector(holdings)\n", " if curr_w.empty:\n", " turnover = 0.0\n", " elif prev_w is None or prev_w.empty:\n", " turnover = curr_w.abs().sum() # initial buy into the portfolio\n", " else:\n", " names = prev_w.index.union(curr_w.index)\n", " turnover = (curr_w.reindex(names, fill_value=0.0) - prev_w.reindex(names, fill_value=0.0)).abs().sum() / 2.0\n", " turnovers.append(float(turnover))\n", " prev_w = curr_w\n", " return pd.Series(turnovers, index=holdings_series.index)\n", "\n", "df_port['long_turnover'] = compute_turnover(df_port['long_holdings'])\n", "df_port['short_turnover'] = compute_turnover(df_port['short_holdings'])\n", "df_port['ls_turnover'] = df_port['long_turnover'] + df_port['short_turnover']\n", "\n", "fig, ax = plt.subplots(figsize=(12, 5))\n", "ax.plot(df_port.index, df_port['long_turnover'], color='steelblue', alpha=0.5, label='Long turnover')\n", "ax.plot(df_port.index, df_port['long_turnover'].rolling(12).mean(), color='coral', linewidth=2, label='12m MA')\n", "ax.set_ylabel('One-Way Turnover')\n", "ax.set_title('Long Portfolio Turnover')\n", "ax.legend()\n", "ax.grid(alpha=0.3)\n", "\n", "plt.tight_layout()\n", "plt.savefig('../images/04_backtest/turnover.png', dpi=150, bbox_inches='tight')\n", "plt.show()\n", "\n", "print(f\"Average long turnover (one-way): {df_port['long_turnover'].mean():.3f}\")\n", "print(f\"Average L/S turnover (one-way): {df_port['ls_turnover'].mean():.3f}\")" ] }, { "cell_type": "code", "execution_count": 6, "id": "d3cfa6a1", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:22.029262Z", "iopub.status.busy": "2026-07-31T12:20:22.029060Z", "iopub.status.idle": "2026-07-31T12:20:22.047668Z", "shell.execute_reply": "2026-07-31T12:20:22.047015Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Net return summary:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
long_netshort_netls_net
count239.0000239.0000239.0000
mean0.01620.0169-0.0007
std0.05600.07000.0479
min-0.1797-0.2208-0.4088
25%-0.0117-0.0209-0.0217
50%0.01740.01440.0019
75%0.04960.04880.0247
max0.17250.47970.1005
\n", "
" ], "text/plain": [ " long_net short_net ls_net\n", "count 239.0000 239.0000 239.0000\n", "mean 0.0162 0.0169 -0.0007\n", "std 0.0560 0.0700 0.0479\n", "min -0.1797 -0.2208 -0.4088\n", "25% -0.0117 -0.0209 -0.0217\n", "50% 0.0174 0.0144 0.0019\n", "75% 0.0496 0.0488 0.0247\n", "max 0.1725 0.4797 0.1005" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\"\"\"\n", "==================================\n", "Apply transaction costs\n", "==================================\n", "\n", "Cost = one-way turnover * cost in bps / 10000.\n", "For the long-short book, long and short costs are applied separately.\n", "\"\"\"\n", "tc = TRANSACTION_COST_BPS / 10000 # 5 bps = 0.0005\n", "\n", "# Net returns = gross - cost\n", "df_port['long_net'] = df_port['long'] - df_port['long_turnover'] * tc\n", "df_port['short_net'] = df_port['short'] + df_port['short_turnover'] * tc # Short pays cost too\n", "df_port['ls_net'] = df_port['long_net'] - df_port['short_net']\n", "\n", "# Excess returns (subtract risk-free)\n", "# Align FF data (month-start) to portfolio index (month-end) via period index\n", "df_port_pm = df_port.copy()\n", "df_port_pm.index = df_port_pm.index.to_period('M')\n", "ff_pm = df_ff[['RF', 'Mkt-RF']].copy()\n", "ff_pm.index = ff_pm.index.to_period('M')\n", "df_port['RF'] = ff_pm['RF'].reindex(df_port_pm.index).values\n", "df_port['Mkt-RF'] = ff_pm['Mkt-RF'].reindex(df_port_pm.index).values\n", "df_port['long_excess'] = df_port['long_net'] - df_port['RF']\n", "df_port['short_excess'] = df_port['short_net'] - df_port['RF']\n", "df_port['ls_excess'] = df_port['ls_net'] # L/S is dollar-neutral, RF cancels\n", "\n", "print(\"Net return summary:\")\n", "display(df_port[['long_net', 'short_net', 'ls_net']].describe().round(4))" ] }, { "cell_type": "markdown", "id": "cb3baab4", "metadata": {}, "source": [ "## Performance Metrics\n", "\n", "The main metrics are:\n", "\n", "- **Sharpe ratio:**\n", "\n", "$$\\text{Sharpe}=\\frac{\\bar r_p}{\\mathrm{std}(r_p)}\\sqrt{12}.$$\n", "\n", "- **Sortino ratio:** like Sharpe, but the denominator is downside deviation rather than total volatility.\n", "- **Max drawdown:** the worst percentage drop from a previous wealth peak:\n", "\n", "$$\\text{DD}_t=\\frac{V_t-\\max_{s\\le t}V_s}{\\max_{s\\le t}V_s}, \\quad V_t=\\prod_{s=1}^{t}(1+r_{p,s}).$$\n", "\n", "- **Calmar ratio:** annualized return divided by the absolute value of max drawdown.\n", "\n", "These are descriptive statistics. They do not prove a strategy is good, but they help us understand the shape of the returns." ] }, { "cell_type": "markdown", "id": "6ab72d21", "metadata": {}, "source": [ "## Sortino Ratio and Max Drawdown\n", "\n", "### Sortino Ratio\n", "\n", "Sharpe uses total volatility in the denominator. That means upside and downside moves both increase the denominator. Sortino replaces total volatility with downside deviation, so months above the target return do not count as risk.\n", "\n", "With target return $r_{\\text{target}}=0$:\n", "\n", "$$\\sigma_D=\\sqrt{\\frac{1}{T}\\sum_{t=1}^{T}\\min(0,r_t-r_{\\text{target}})^2},$$\n", "\n", "and\n", "\n", "$$\\text{Sortino}=\\frac{\\bar r_p}{\\sigma_D}\\sqrt{12}.$$\n", "\n", "Example monthly returns:\n", "\n", "$$[0.03,-0.02,0.08,-0.01,0.04].$$\n", "\n", "Only the negative months enter downside deviation:\n", "\n", "| Month | Return | Downside part | Squared |\n", "|---|---:|---:|---:|\n", "| 1 | 0.03 | 0.00 | 0.0000 |\n", "| 2 | -0.02 | -0.02 | 0.0004 |\n", "| 3 | 0.08 | 0.00 | 0.0000 |\n", "| 4 | -0.01 | -0.01 | 0.0001 |\n", "| 5 | 0.04 | 0.00 | 0.0000 |\n", "\n", "So\n", "\n", "$$\\sigma_D=\\sqrt{(0.0004+0.0001)/5}=0.01.$$\n", "\n", "Sortino is useful when the return distribution is asymmetric. A large gap between Sortino and Sharpe usually means the strategy has more upside volatility than downside volatility.\n", "\n", "### Max Drawdown\n", "\n", "Cumulative wealth is\n", "\n", "$$V_t=(1+r_1)(1+r_2)\\cdots(1+r_t).$$\n", "\n", "The running peak is\n", "\n", "$$P_t=\\max_{s\\le t}V_s.$$\n", "\n", "Drawdown is the percentage distance below that peak:\n", "\n", "$$\\text{DD}_t=\\frac{V_t-P_t}{P_t}.$$\n", "\n", "Example:\n", "\n", "| Month | Wealth $V_t$ | Running peak $P_t$ | Drawdown |\n", "|---|---:|---:|---:|\n", "| 1 | 1.05 | 1.05 | 0.0% |\n", "| 2 | 1.08 | 1.08 | 0.0% |\n", "| 3 | 1.02 | 1.08 | -5.6% |\n", "| 4 | 0.95 | 1.08 | -12.0% |\n", "| 5 | 1.01 | 1.08 | -6.5% |\n", "\n", "Max drawdown is the worst value in that drawdown series. It answers: how far underwater would an investor have been at the worst point?\n", "\n", "Calmar then asks how much annual return the strategy earned per unit of that worst drawdown:\n", "\n", "$$\\text{Calmar}=\\frac{\\text{annualized return}}{|\\text{max drawdown}|}.$$" ] }, { "cell_type": "markdown", "id": "d2886153", "metadata": {}, "source": [ "### Equal-Weight (EW) Universe Benchmark\n", "\n", "The equal-weight universe holds every available stock at the same weight:\n", "\n", "$$w_i=\\frac{1}{N_t}.$$\n", "\n", "Its return is the row mean of the return matrix:\n", "\n", "$$\\bar r_t=\\frac{1}{N_t}\\mathbf{1}^\\top r_t.$$\n", "\n", "This is a better benchmark for this project than a cap-weighted index because our portfolio is also equal-weighted within its selected names. Comparing equal-weight to equal-weight keeps the focus on selection: did the top-momentum decile beat simply holding the whole available universe equally?\n", "\n", "The difference $r_{p,t}-\\bar r_t$ is the **active return**, and its annualized mean divided by tracking error is the **information ratio**." ] }, { "cell_type": "code", "execution_count": 7, "id": "4845c9f1", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:22.049879Z", "iopub.status.busy": "2026-07-31T12:20:22.049541Z", "iopub.status.idle": "2026-07-31T12:20:22.070626Z", "shell.execute_reply": "2026-07-31T12:20:22.069944Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Performance Summary\n", "\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
ann_returnann_volsharpesortinomax_drawdowncalmar
long_net0.19480.19411.00361.6436-0.57130.3409
ls_net-0.00810.1658-0.0491-0.0590-0.7029-0.0116
ew_universe0.15940.16830.94711.5203-0.47470.3359
\n", "
" ], "text/plain": [ " ann_return ann_vol sharpe sortino max_drawdown calmar\n", "long_net 0.1948 0.1941 1.0036 1.6436 -0.5713 0.3409\n", "ls_net -0.0081 0.1658 -0.0491 -0.0590 -0.7029 -0.0116\n", "ew_universe 0.1594 0.1683 0.9471 1.5203 -0.4747 0.3359" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "Active Return (Long-Only minus EW Universe)\n", " Monthly active return: 0.00295 (t = 1.89)\n", " Annualized: 0.0354\n", " Tracking error: 0.0833\n", " Information ratio: 0.4243\n" ] } ], "source": [ "\"\"\"\n", "==================================\n", "Compute performance metrics\n", "==================================\n", "\"\"\"\n", "def performance_metrics(returns, freq=12, rf=0):\n", " \"\"\"Compute standard performance metrics.\"\"\"\n", " returns = returns.dropna()\n", " excess = returns - rf\n", " ann_return = returns.mean() * freq\n", " ann_vol = returns.std() * np.sqrt(freq)\n", " sharpe = ann_return / ann_vol if ann_vol > 0 else np.nan\n", "\n", " downside = np.minimum(excess, 0.0)\n", " downside_vol = np.sqrt(np.mean(downside ** 2)) * np.sqrt(freq) if len(downside) > 0 else np.nan\n", " sortino = ann_return / downside_vol if downside_vol and downside_vol > 0 else np.nan\n", "\n", " cum = (1 + returns).cumprod()\n", " running_max = cum.cummax()\n", " drawdown = (cum - running_max) / running_max\n", " max_dd = drawdown.min()\n", " \n", " calmar = ann_return / abs(max_dd) if max_dd < 0 else np.nan\n", " \n", " return {\n", " 'ann_return': ann_return,\n", " 'ann_vol': ann_vol,\n", " 'sharpe': sharpe,\n", " 'sortino': sortino,\n", " 'max_drawdown': max_dd,\n", " 'calmar': calmar\n", " }\n", "\n", "# Equal-weight universe benchmark — align via period index\n", "ew_monthly = df_returns.mean(axis=1)\n", "ew_monthly.index = ew_monthly.index.to_period('M')\n", "df_port['ew_universe'] = ew_monthly.reindex(df_port.index.to_period('M')).values\n", "\n", "# Compute metrics for all portfolios\n", "metrics = {}\n", "for col, label in [('long_net', 'Long-Only (net)'), ('ls_net', 'Long-Short (net)'), ('ew_universe', 'EW Universe')]:\n", " metrics[col] = performance_metrics(df_port[col])\n", "\n", "df_metrics = pd.DataFrame(metrics).T.round(4)\n", "print(\"Performance Summary\\n\")\n", "display(df_metrics)\n", "\n", "# Active return vs equal-weight universe\n", "df_port['active'] = df_port['long_net'] - df_port['ew_universe']\n", "active = df_port['active'].dropna()\n", "t_stat = active.mean() / (active.std() / np.sqrt(len(active)))\n", "ir = active.mean() / active.std() * np.sqrt(12)\n", "print(f\"\\nActive Return (Long-Only minus EW Universe)\")\n", "print(f\" Monthly active return: {active.mean():.5f} (t = {t_stat:.2f})\")\n", "print(f\" Annualized: {active.mean()*12:.4f}\")\n", "print(f\" Tracking error: {active.std()*np.sqrt(12):.4f}\")\n", "print(f\" Information ratio: {ir:.4f}\")" ] }, { "cell_type": "code", "execution_count": 8, "id": "e9999e92", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:22.073634Z", "iopub.status.busy": "2026-07-31T12:20:22.073011Z", "iopub.status.idle": "2026-07-31T12:20:22.927406Z", "shell.execute_reply": "2026-07-31T12:20:22.926772Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\"\"\"\n", "==================================\n", "Equity curve and drawdown\n", "==================================\n", "\"\"\"\n", "fig, axes = plt.subplots(2, 1, figsize=(14, 10), sharex=True)\n", "\n", "# Cumulative returns\n", "ax = axes[0]\n", "for col, color, label in [('long_net', 'steelblue', 'Long-Only'), ('ls_net', 'coral', 'Long-Short')]:\n", " cum = (1 + df_port[col]).cumprod()\n", " ax.plot(cum.index, cum.values, color=color, linewidth=2, label=label)\n", "ax.set_ylabel('Cumulative Return')\n", "ax.set_title('Portfolio Equity Curves (Net of Costs)')\n", "ax.legend()\n", "ax.grid(alpha=0.3)\n", "\n", "# Drawdown\n", "ax = axes[1]\n", "for col, color, label in [('long_net', 'steelblue', 'Long-Only'), ('ls_net', 'coral', 'Long-Short')]:\n", " cum = (1 + df_port[col]).cumprod()\n", " dd = (cum - cum.cummax()) / cum.cummax()\n", " ax.fill_between(dd.index, dd.values, 0, color=color, alpha=0.3, label=label)\n", "ax.set_ylabel('Drawdown')\n", "ax.set_title('Portfolio Drawdowns')\n", "ax.legend()\n", "ax.grid(alpha=0.3)\n", "\n", "plt.tight_layout()\n", "plt.savefig('../images/04_backtest/equity_and_drawdown.png', dpi=150, bbox_inches='tight')\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "54006d1a", "metadata": {}, "source": [ "## Walk-Forward Analysis\n", "\n", "A full-sample Sharpe can be misleading if one period did all the work. Walk-forward analysis splits the history into 5-year windows and recomputes performance in each window.\n", "\n", "This is a consistency check, not a parameter search. The question is whether the strategy works across more than one regime." ] }, { "cell_type": "code", "execution_count": 9, "id": "a0a6cf4f", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:22.929561Z", "iopub.status.busy": "2026-07-31T12:20:22.929356Z", "iopub.status.idle": "2026-07-31T12:20:23.333476Z", "shell.execute_reply": "2026-07-31T12:20:23.332807Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Walk-Forward Performance (5-Year Windows)\n", "\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
lo_sharpelo_ann_retls_sharpeew_sharpeactive_retinfo_ratioactive_t
period
2006-20110.31080.0703-0.54930.5729-0.0495-0.5230-1.1596
2011-20161.43470.21050.66021.34700.04200.78021.7445
2016-20211.28100.2515-0.00181.11720.05920.79941.7874
2021-20261.22790.24470.29420.97380.08830.88521.9795
\n", "
" ], "text/plain": [ " lo_sharpe lo_ann_ret ls_sharpe ew_sharpe active_ret \\\n", "period \n", "2006-2011 0.3108 0.0703 -0.5493 0.5729 -0.0495 \n", "2011-2016 1.4347 0.2105 0.6602 1.3470 0.0420 \n", "2016-2021 1.2810 0.2515 -0.0018 1.1172 0.0592 \n", "2021-2026 1.2279 0.2447 0.2942 0.9738 0.0883 \n", "\n", " info_ratio active_t \n", "period \n", "2006-2011 -0.5230 -1.1596 \n", "2011-2016 0.7802 1.7445 \n", "2016-2021 0.7994 1.7874 \n", "2021-2026 0.8852 1.9795 " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "Long-Only Sharpe positive in 4/4 windows\n", "Active return positive in 3/4 windows\n" ] } ], "source": [ "\"\"\"\n", "==================================\n", "Walk-forward analysis: 5-year windows\n", "==================================\n", "\"\"\"\n", "windows = [(2006, 2011), (2011, 2016), (2016, 2021), (2021, 2026)]\n", "\n", "wf_results = []\n", "for start, end in windows:\n", " mask = (df_port.index.year >= start) & (df_port.index.year < end)\n", " if mask.sum() < 12:\n", " continue\n", " \n", " m_lo = performance_metrics(df_port.loc[mask, 'long_net'])\n", " m_ls = performance_metrics(df_port.loc[mask, 'ls_net'])\n", " m_ew = performance_metrics(df_port.loc[mask, 'ew_universe'])\n", " \n", " active_sub = df_port.loc[mask, 'active'].dropna()\n", " ir = active_sub.mean() / active_sub.std() * np.sqrt(12) if len(active_sub) > 12 and active_sub.std() > 0 else np.nan\n", " t_act = active_sub.mean() / (active_sub.std() / np.sqrt(len(active_sub))) if len(active_sub) > 12 else np.nan\n", " \n", " wf_results.append({\n", " 'period': f'{start}-{end}',\n", " 'lo_sharpe': m_lo['sharpe'],\n", " 'lo_ann_ret': m_lo['ann_return'],\n", " 'ls_sharpe': m_ls['sharpe'],\n", " 'ew_sharpe': m_ew['sharpe'],\n", " 'active_ret': active_sub.mean() * 12,\n", " 'info_ratio': ir,\n", " 'active_t': t_act\n", " })\n", "\n", "df_wf = pd.DataFrame(wf_results).set_index('period')\n", "print(\"Walk-Forward Performance (5-Year Windows)\\n\")\n", "display(df_wf.round(4))\n", "\n", "# Plot walk-forward Sharpe ratios\n", "fig, ax = plt.subplots(figsize=(10, 6))\n", "x = np.arange(len(df_wf))\n", "width = 0.25\n", "ax.bar(x - width, df_wf['lo_sharpe'], width, label='Long-Only', color='steelblue')\n", "ax.bar(x, df_wf['ew_sharpe'], width, label='EW Universe', color='coral')\n", "ax.bar(x + width, df_wf['ls_sharpe'], width, label='Long-Short', color='seagreen')\n", "ax.set_xticks(x)\n", "ax.set_xticklabels(df_wf.index)\n", "ax.set_ylabel('Sharpe Ratio')\n", "ax.set_title('Walk-Forward Sharpe by Subperiod')\n", "ax.axhline(y=0, color='black', linewidth=0.5)\n", "ax.legend()\n", "ax.grid(alpha=0.3, axis='y')\n", "\n", "plt.tight_layout()\n", "plt.savefig('../images/04_backtest/walk_forward.png', dpi=150, bbox_inches='tight')\n", "plt.show()\n", "\n", "print(f\"\\nLong-Only Sharpe positive in {(df_wf['lo_sharpe'] > 0).sum()}/{len(df_wf)} windows\")\n", "print(f\"Active return positive in {(df_wf['active_ret'] > 0).sum()}/{len(df_wf)} windows\")" ] }, { "cell_type": "markdown", "id": "14f058aa", "metadata": {}, "source": [ "## Reconciling Walk-Forward IC vs. Walk-Forward Sharpe\n", "\n", "Notebook 02 found that momentum's monthly IC was positive in only some subperiods, while the long-only portfolio Sharpe can still be positive across all windows. Those are not contradictory.\n", "\n", "IC measures cross-sectional ordering: did higher-ranked stocks beat lower-ranked stocks? Portfolio Sharpe measures the return level of the selected stocks. A top-decile book can make money in a window even if its ranking skill is weak, especially if the selected stocks had high market beta during a rising market.\n", "\n", "The table below compares mean IC, portfolio Sharpe, and realized market beta by window. If high beta lines up with weak IC, the Sharpe is probably being carried by market exposure rather than stock selection." ] }, { "cell_type": "code", "execution_count": 10, "id": "508e0b07", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:23.335403Z", "iopub.status.busy": "2026-07-31T12:20:23.335208Z", "iopub.status.idle": "2026-07-31T12:20:23.360707Z", "shell.execute_reply": "2026-07-31T12:20:23.360069Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "IC vs. Sharpe vs. realized market beta, by walk-forward window:\n", "\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
mean_icportfolio_sharpeportfolio_mkt_beta
period
2006-2011-0.0130.3111.144
2011-20160.0281.4351.105
2016-2021-0.0081.2811.127
2021-20260.0171.2281.171
\n", "
" ], "text/plain": [ " mean_ic portfolio_sharpe portfolio_mkt_beta\n", "period \n", "2006-2011 -0.013 0.311 1.144\n", "2011-2016 0.028 1.435 1.105\n", "2016-2021 -0.008 1.281 1.127\n", "2021-2026 0.017 1.228 1.171" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "If portfolio_mkt_beta is visibly higher in the windows where mean_ic is negative, that is the mechanical explanation: the decile's Sharpe in that window is being carried by market exposure rather than by the ranking signal actually working that period.\n" ] } ], "source": [ "\"\"\"\n", "==================================\n", "Reconcile factor-level IC (notebook 02) with portfolio-level Sharpe, by window\n", "==================================\n", "\"\"\"\n", "df_ic_momentum = pd.read_csv('../data/processed/ic_monthly.csv', index_col=0, parse_dates=True)['momentum']\n", "\n", "reconcile_rows = []\n", "for start, end in windows:\n", " mask_port = (df_port.index.year >= start) & (df_port.index.year < end)\n", " mask_ic = (df_ic_momentum.index.year >= start) & (df_ic_momentum.index.year < end)\n", "\n", " sub_port = df_port.loc[mask_port].dropna(subset=['long_excess', 'Mkt-RF'])\n", " sub_ic = df_ic_momentum.loc[mask_ic].dropna()\n", "\n", " if len(sub_port) > 3 and sub_port['Mkt-RF'].var() > 0:\n", " port_beta = sub_port['long_excess'].cov(sub_port['Mkt-RF']) / sub_port['Mkt-RF'].var()\n", " else:\n", " port_beta = np.nan\n", "\n", " reconcile_rows.append({\n", " 'period': f'{start}-{end}',\n", " 'mean_ic': sub_ic.mean(),\n", " 'portfolio_sharpe': performance_metrics(df_port.loc[mask_port, 'long_net'])['sharpe'],\n", " 'portfolio_mkt_beta': port_beta,\n", " })\n", "\n", "df_reconcile = pd.DataFrame(reconcile_rows).set_index('period')\n", "print(\"IC vs. Sharpe vs. realized market beta, by walk-forward window:\\n\")\n", "display(df_reconcile.round(3))\n", "\n", "print(\n", " \"\\nIf portfolio_mkt_beta is visibly higher in the windows where mean_ic is negative, \"\n", " \"that is the mechanical explanation: the decile's Sharpe in that window is being carried \"\n", " \"by market exposure rather than by the ranking signal actually working that period.\"\n", ")\n" ] }, { "cell_type": "markdown", "id": "79a4d8c5", "metadata": {}, "source": [ "## Fama-French Alpha\n", "\n", "The Fama-French regression is an OLS regression of portfolio excess returns on benchmark factor returns. Given factor matrix $F \\in \\mathbb{R}^{T\\times k}$ and portfolio excess returns $y\\in\\mathbb{R}^T$, the fitted model is\n", "\n", "$$y=\\alpha\\mathbf{1}+F\\beta+\\varepsilon.$$\n", "\n", "The least-squares beta estimate is\n", "\n", "$$\\hat\\beta=(F^\\top F)^{-1}F^\\top(y-\\alpha\\mathbf{1}),$$\n", "\n", "or equivalently the full coefficient vector is estimated after adding a column of ones to $F$.\n", "\n", "- **Betas:** exposures to benchmark factors such as market, size, value, and momentum.\n", "- **Alpha:** the intercept. It is the average return left after controlling for those factor exposures.\n", "- **Residuals:** the month-by-month unexplained returns around the fitted line.\n", "\n", "So alpha is related to the residual, but it is not the entire residual vector. It is the average unexplained return, annualized here by multiplying the monthly intercept by 12.\n", "\n", "We report the usual OLS t-statistic and a **HAC/Newey-West** t-statistic with three monthly lags. HAC standard errors are a useful check because monthly portfolio residuals can have mild autocorrelation or changing volatility. If the alpha only survives under plain OLS and disappears under HAC, the result is less convincing.\n", "\n", "We run this on the long-only portfolio as the headline result and on the long-short portfolio as a comparison." ] }, { "cell_type": "code", "execution_count": 11, "id": "10c3bb8c", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:23.362870Z", "iopub.status.busy": "2026-07-31T12:20:23.362667Z", "iopub.status.idle": "2026-07-31T12:20:23.581378Z", "shell.execute_reply": "2026-07-31T12:20:23.580746Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Regression data points: 239\n", "============================================================\n", "LONG-ONLY: Fama-French 4-Factor Alpha\n", "============================================================\n", " Alpha (monthly): 0.00497\n", " Alpha (annualized): 0.0597\n", " Alpha t-stat (OLS): 3.98\n", " Alpha t-stat (HAC): 4.17 (Newey-West, 3 lags)\n", " MKT beta: 1.189 (t=39.27)\n", " SMB beta: 0.262 (t=5.02)\n", " HML beta: -0.045 (t=-1.12)\n", " MOM beta: 0.251 (t=7.95)\n", " R^2: 0.889\n", "\n", "============================================================\n", "LONG-SHORT: Fama-French 4-Factor Alpha (comparison)\n", "============================================================\n", " Alpha (monthly): -0.00246\n", " Alpha (annualized): -0.0295\n", " Alpha t-stat (OLS): -1.41\n", " Alpha t-stat (HAC): -1.37 (Newey-West, 3 lags)\n", " MKT beta: 0.144 (t=3.42)\n", " MOM beta: 0.905 (t=20.62)\n", " R^2: 0.704\n", "\n", "============================================================\n", "SUMMARY\n", " Long-only alpha: +0.0597 (OLS t=3.98, HAC t=4.17)\n", " Long-short alpha: -0.0295 (OLS t=-1.41, HAC t=-1.37)\n", "============================================================\n" ] } ], "source": [ "\"\"\"\n", "==================================\n", "Fama-French 4-factor regression: Long-Only and Long-Short\n", "==================================\n", "\"\"\"\n", "import statsmodels.api as sm\n", "\n", "FF_FACTOR_COLS = ['Mkt-RF', 'SMB', 'HML', 'Mom']\n", "HAC_LAGS = 3\n", "\n", "# --- Robust alignment via year-month period index ---\n", "# df_port has month-end dates, df_ff has month-start dates.\n", "# Align both to PeriodIndex('M') so they match regardless of timestamp conventions.\n", "df_port_pm = df_port.copy()\n", "df_port_pm.index = df_port_pm.index.to_period('M')\n", "\n", "df_ff_pm = df_ff[FF_FACTOR_COLS].copy()\n", "df_ff_pm.index = df_ff_pm.index.to_period('M')\n", "\n", "df_reg = df_port_pm.join(df_ff_pm, how='inner', lsuffix='_port', rsuffix='_ff')\n", "\n", "# Resolve any column-name collisions (Mkt-RF may exist in both from cell 10)\n", "for col in FF_FACTOR_COLS:\n", " if col + '_ff' in df_reg.columns:\n", " df_reg[col] = df_reg[col + '_ff']\n", "\n", "df_reg = df_reg[['long_excess', 'ls_net'] + FF_FACTOR_COLS].dropna().astype(float)\n", "\n", "print(f\"Regression data points: {len(df_reg)}\")\n", "\n", "\n", "def fit_ff_model(y):\n", " \"\"\"Fit FF regression with ordinary and HAC/Newey-West covariance.\"\"\"\n", " X = sm.add_constant(df_reg[FF_FACTOR_COLS], has_constant='add')\n", " ols = sm.OLS(y, X).fit()\n", " hac = sm.OLS(y, X).fit(cov_type='HAC', cov_kwds={'maxlags': HAC_LAGS})\n", " return ols, hac\n", "\n", "\n", "def print_ff_model(label, model, model_hac, show_all_betas=True):\n", " print(\"=\" * 60)\n", " print(label)\n", " print(\"=\" * 60)\n", " print(f\" Alpha (monthly): {model.params['const']:.5f}\")\n", " print(f\" Alpha (annualized): {model.params['const']*12:.4f}\")\n", " print(f\" Alpha t-stat (OLS): {model.tvalues['const']:.2f}\")\n", " print(f\" Alpha t-stat (HAC): {model_hac.tvalues['const']:.2f} (Newey-West, {HAC_LAGS} lags)\")\n", " print(f\" MKT beta: {model.params['Mkt-RF']:.3f} (t={model.tvalues['Mkt-RF']:.2f})\")\n", " if show_all_betas:\n", " print(f\" SMB beta: {model.params['SMB']:.3f} (t={model.tvalues['SMB']:.2f})\")\n", " print(f\" HML beta: {model.params['HML']:.3f} (t={model.tvalues['HML']:.2f})\")\n", " print(f\" MOM beta: {model.params['Mom']:.3f} (t={model.tvalues['Mom']:.2f})\")\n", " else:\n", " print(f\" MOM beta: {model.params['Mom']:.3f} (t={model.tvalues['Mom']:.2f})\")\n", " print(f\" R^2: {model.rsquared:.3f}\")\n", "\n", "if len(df_reg) == 0:\n", " print(\"ERROR: No overlapping data between backtest and Fama-French factors.\")\n", " print(\"Check the indices of df_port and df_ff.\")\n", "else:\n", " model_lo, model_lo_hac = fit_ff_model(df_reg['long_excess'])\n", " print_ff_model(\"LONG-ONLY: Fama-French 4-Factor Alpha\", model_lo, model_lo_hac)\n", "\n", " model_ls, model_ls_hac = fit_ff_model(df_reg['ls_net'])\n", " print()\n", " print_ff_model(\"LONG-SHORT: Fama-French 4-Factor Alpha (comparison)\", model_ls, model_ls_hac, show_all_betas=False)\n", "\n", " print(f\"\\n{'=' * 60}\")\n", " print(\"SUMMARY\")\n", " print(f\" Long-only alpha: {model_lo.params['const']*12:+.4f} \"\n", " f\"(OLS t={model_lo.tvalues['const']:.2f}, HAC t={model_lo_hac.tvalues['const']:.2f})\")\n", " print(f\" Long-short alpha: {model_ls.params['const']*12:+.4f} \"\n", " f\"(OLS t={model_ls.tvalues['const']:.2f}, HAC t={model_ls_hac.tvalues['const']:.2f})\")\n", " print(\"=\" * 60)" ] }, { "cell_type": "markdown", "id": "0a7da761", "metadata": {}, "source": [ "### What Does MKT Beta = 1.19 Mean for the Headline Alpha?\n", "\n", "The regression reports a market beta along with alpha. A beta around 1.19 means the long-only decile had about 19% more market exposure than a beta-1 portfolio over this sample.\n", "\n", "That matters because part of the raw return may be ordinary market risk, not stock selection. This is why the factor-adjusted alpha is more informative than the raw active return. The regression subtracts the part explained by market, size, value, and momentum exposure before estimating the intercept.\n", "\n", "To make the idea concrete, we compare the portfolio to a beta-matched benchmark: the equal-weight universe scaled to the same market beta. This is not a tradable recommendation; it is a diagnostic for whether the outperformance survives a simple market-risk adjustment." ] }, { "cell_type": "code", "execution_count": 12, "id": "519db558", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:23.583614Z", "iopub.status.busy": "2026-07-31T12:20:23.583280Z", "iopub.status.idle": "2026-07-31T12:20:23.590261Z", "shell.execute_reply": "2026-07-31T12:20:23.589581Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Long-only ann. return: 0.1948\n", "EW universe ann. return (unscaled): 0.1594\n", "EW universe ann. return (x1.19 beta-matched): 0.1895\n", "\n", "Outperformance vs. unscaled EW: +0.0354\n", "Outperformance vs. beta-matched EW: +0.0052\n", "\n", "If the beta-matched gap is still clearly positive, the outperformance is not simply leverage on the market factor; this is a second, more direct check on the same question the FF alpha answers via regression.\n" ] } ], "source": [ "\"\"\"\n", "==================================\n", "Beta-matched benchmark comparison\n", "==================================\n", "\"\"\"\n", "beta_lo = model_lo.params['Mkt-RF'] # MKT-RF coefficient from the FF regression above\n", "\n", "ew_scaled = df_port['ew_universe'] * beta_lo\n", "\n", "lo_ann_ret = df_port['long_net'].mean() * 12\n", "ew_ann_ret = df_port['ew_universe'].mean() * 12\n", "ew_scaled_ann_ret = ew_scaled.mean() * 12\n", "\n", "print(f\"Long-only ann. return: {lo_ann_ret:.4f}\")\n", "print(f\"EW universe ann. return (unscaled): {ew_ann_ret:.4f}\")\n", "print(f\"EW universe ann. return (x{beta_lo:.2f} beta-matched): {ew_scaled_ann_ret:.4f}\")\n", "print(f\"\\nOutperformance vs. unscaled EW: {lo_ann_ret - ew_ann_ret:+.4f}\")\n", "print(f\"Outperformance vs. beta-matched EW: {lo_ann_ret - ew_scaled_ann_ret:+.4f}\")\n", "print(\n", " \"\\nIf the beta-matched gap is still clearly positive, the outperformance is not simply \"\n", " \"leverage on the market factor; this is a second, more direct check on the same question \"\n", " \"the FF alpha answers via regression.\"\n", ")\n" ] }, { "cell_type": "markdown", "id": "b8020d29", "metadata": {}, "source": [ "## Survivorship Bias Sensitivity\n", "\n", "The universe is based on current S&P 500 constituents. That means stocks that disappeared from the index, were acquired, delisted, or went bankrupt may be missing from the historical panel.\n", "\n", "We cannot fully fix this without survivorship-free data. Instead, we ask a sensitivity question: how much annual return drag would we need to subtract before the Fama-French alpha is no longer statistically significant?\n", "\n", "This is not a perfect model of delisting bias. It is a breakeven calculation. If a small drag erases the result, the backtest is fragile. If a large drag is needed, the result is less likely to be explained only by survivorship bias." ] }, { "cell_type": "code", "execution_count": 13, "id": "5a4bcd3f", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:23.592778Z", "iopub.status.busy": "2026-07-31T12:20:23.592552Z", "iopub.status.idle": "2026-07-31T12:20:23.971023Z", "shell.execute_reply": "2026-07-31T12:20:23.970290Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "### Survivorship Bias Sensitivity\n", "\n", "How much annual return drag from missing delisted stocks would erase the alpha?\n", "\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
annual_dragalpha_monthlyalpha_annualizedt_stat_olst_stat_hacsignificant_hac
00.0%0.00500.05973.97774.1656Yes
10.5%0.00460.05473.64443.8166Yes
21.0%0.00410.04973.31113.4676Yes
32.0%0.00330.03972.64462.7695Yes
43.0%0.00250.02971.97802.0714Yes
54.0%0.00160.01971.31141.3734No
65.0%0.00080.00970.64490.6753No
\n", "
" ], "text/plain": [ " annual_drag alpha_monthly alpha_annualized t_stat_ols t_stat_hac \\\n", "0 0.0% 0.0050 0.0597 3.9777 4.1656 \n", "1 0.5% 0.0046 0.0547 3.6444 3.8166 \n", "2 1.0% 0.0041 0.0497 3.3111 3.4676 \n", "3 2.0% 0.0033 0.0397 2.6446 2.7695 \n", "4 3.0% 0.0025 0.0297 1.9780 2.0714 \n", "5 4.0% 0.0016 0.0197 1.3114 1.3734 \n", "6 5.0% 0.0008 0.0097 0.6449 0.6753 \n", "\n", " significant_hac \n", "0 Yes \n", "1 Yes \n", "2 Yes \n", "3 Yes \n", "4 Yes \n", "5 No \n", "6 No " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\"\"\"\n", "==================================\n", "Survivorship bias sensitivity\n", "==================================\n", "\"\"\"\n", "drags = [0, 0.005, 0.01, 0.02, 0.03, 0.04, 0.05]\n", "sensitivity = []\n", "\n", "# Reuse the aligned FF regression frame built in the alpha cell;\n", "# rebuild it if this cell is run standalone.\n", "if 'df_reg' not in globals():\n", " _pm = df_port.copy(); _pm.index = _pm.index.to_period('M')\n", " _ff = df_ff[['Mkt-RF', 'SMB', 'HML', 'Mom']].copy(); _ff.index = _ff.index.to_period('M')\n", " df_reg = _pm.join(_ff, how='inner', lsuffix='_port', rsuffix='_ff')\n", " for _c in ['Mkt-RF', 'SMB', 'HML', 'Mom']:\n", " if _c + '_ff' in df_reg.columns:\n", " df_reg[_c] = df_reg[_c + '_ff']\n", " df_reg = df_reg[['long_excess', 'Mkt-RF', 'SMB', 'HML', 'Mom']].dropna().astype(float)\n", "\n", "X_factors = sm.add_constant(df_reg[['Mkt-RF', 'SMB', 'HML', 'Mom']])\n", "\n", "for drag in drags:\n", " y_adj = df_reg['long_excess'] - drag / 12 # monthly drag on excess returns\n", " model = sm.OLS(y_adj, X_factors).fit()\n", " model_hac = sm.OLS(y_adj, X_factors).fit(cov_type='HAC', cov_kwds={'maxlags': HAC_LAGS})\n", " alpha_m = model.params['const']\n", " alpha_t = model.tvalues['const']\n", " sensitivity.append({\n", " 'annual_drag': f'{drag*100:.1f}%',\n", " 'alpha_monthly': alpha_m,\n", " 'alpha_annualized': alpha_m * 12,\n", " 't_stat_ols': alpha_t,\n", " 't_stat_hac': model_hac.tvalues['const'],\n", " 'significant_hac': 'Yes' if abs(model_hac.tvalues['const']) > 2 else 'No'\n", " })\n", "\n", "df_sens = pd.DataFrame(sensitivity)\n", "print(\"### Survivorship Bias Sensitivity\\n\")\n", "print(\"How much annual return drag from missing delisted stocks would erase the alpha?\\n\")\n", "display(df_sens.round(4))\n", "\n", "fig, ax = plt.subplots(figsize=(10, 5))\n", "ax.plot(df_sens['annual_drag'], df_sens['t_stat_ols'], 'o-', color='steelblue', linewidth=2, label='OLS t-stat')\n", "ax.plot(df_sens['annual_drag'], df_sens['t_stat_hac'], 'o-', color='seagreen', linewidth=2, label='HAC t-stat')\n", "ax.axhline(y=2, color='coral', linestyle='--', label='t = 2 (5% significance)')\n", "ax.axhline(y=0, color='black', linewidth=0.5)\n", "ax.set_xlabel('Synthetic Annual Return Drag')\n", "ax.set_ylabel('FF 4-Factor Alpha t-Statistic')\n", "ax.set_title('Survivorship Bias Sensitivity: When Does Alpha Disappear?')\n", "ax.legend()\n", "ax.grid(alpha=0.3)\n", "\n", "plt.tight_layout()\n", "plt.savefig('../images/04_backtest/survivorship_sensitivity.png', dpi=150, bbox_inches='tight')\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "2b3dbdb0", "metadata": {}, "source": [ "### Concentrating the Drag in Bad Months\n", "\n", "A flat monthly drag is easy to read, but real delisting losses are not evenly spread through time. They tend to cluster during market stress.\n", "\n", "So we run a harsher version: apply most of the same annual drag during the worst equal-weight universe months and only a small residual drag elsewhere. If this concentrated drag erases the alpha much faster, the flat-drag test was too generous." ] }, { "cell_type": "code", "execution_count": 14, "id": "1a9a235a", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:23.973287Z", "iopub.status.busy": "2026-07-31T12:20:23.973043Z", "iopub.status.idle": "2026-07-31T12:20:24.014214Z", "shell.execute_reply": "2026-07-31T12:20:24.013558Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Concentrated (crash-weighted) drag sensitivity:\n", "\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
annual_dragalpha_annualizedalpha_t_olsalpha_t_hacsignificant_hac_t_gt_2
00.0%0.05973.97774.1656True
10.5%0.05383.59213.7664True
21.0%0.04793.20423.3635True
32.0%0.03622.42272.5481True
43.0%0.02441.63641.7237False
54.0%0.01270.84840.8946False
65.0%0.00090.06180.0652False
\n", "
" ], "text/plain": [ " annual_drag alpha_annualized alpha_t_ols alpha_t_hac \\\n", "0 0.0% 0.0597 3.9777 4.1656 \n", "1 0.5% 0.0538 3.5921 3.7664 \n", "2 1.0% 0.0479 3.2042 3.3635 \n", "3 2.0% 0.0362 2.4227 2.5481 \n", "4 3.0% 0.0244 1.6364 1.7237 \n", "5 4.0% 0.0127 0.8484 0.8946 \n", "6 5.0% 0.0009 0.0618 0.0652 \n", "\n", " significant_hac_t_gt_2 \n", "0 True \n", "1 True \n", "2 True \n", "3 True \n", "4 False \n", "5 False \n", "6 False " ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "Compare the drag level at which significance is lost here against the flat-drag table above. A meaningfully lower breakeven drag under concentration would mean the flat-drag result overstates how robust the alpha is to realistic (crash-clustered) survivorship bias.\n" ] } ], "source": [ "\"\"\"\n", "==================================\n", "Concentrated (crash-weighted) survivorship drag sensitivity\n", "==================================\n", "\"\"\"\n", "crash_mask = df_reg.index.map(\n", " lambda p: df_port_pm.loc[p, 'ew_universe'] if p in df_port_pm.index else np.nan\n", ")\n", "crash_mask = pd.Series(crash_mask, index=df_reg.index)\n", "crash_threshold = crash_mask.quantile(0.25)\n", "is_crash_month = crash_mask < crash_threshold\n", "\n", "concentrated_sensitivity = []\n", "for drag in drags:\n", " monthly_drag = drag / 12\n", " y_adj = df_reg['long_excess'].copy()\n", " # ~80% of the total drag falls in the worst quartile of months, 20% spread over the rest\n", " n_crash = is_crash_month.sum()\n", " n_other = (~is_crash_month).sum()\n", " if n_crash > 0:\n", " y_adj[is_crash_month] -= monthly_drag * len(y_adj) * 0.8 / n_crash\n", " if n_other > 0:\n", " y_adj[~is_crash_month] -= monthly_drag * len(y_adj) * 0.2 / n_other\n", "\n", " model = sm.OLS(y_adj, X_factors).fit()\n", " model_hac = sm.OLS(y_adj, X_factors).fit(cov_type='HAC', cov_kwds={'maxlags': HAC_LAGS})\n", " alpha_m = model.params['const']\n", " concentrated_sensitivity.append({\n", " 'annual_drag': f'{drag*100:.1f}%',\n", " 'alpha_annualized': alpha_m * 12,\n", " 'alpha_t_ols': model.tvalues['const'],\n", " 'alpha_t_hac': model_hac.tvalues['const'],\n", " 'significant_hac_t_gt_2': abs(model_hac.tvalues['const']) > 2,\n", " })\n", "\n", "df_concentrated_sensitivity = pd.DataFrame(concentrated_sensitivity)\n", "print(\"Concentrated (crash-weighted) drag sensitivity:\\n\")\n", "display(df_concentrated_sensitivity.round(4))\n", "print(\n", " \"\\nCompare the drag level at which significance is lost here against the flat-drag table above. \"\n", " \"A meaningfully lower breakeven drag under concentration would mean the flat-drag result \"\n", " \"overstates how robust the alpha is to realistic (crash-clustered) survivorship bias.\"\n", ")\n" ] }, { "cell_type": "code", "execution_count": 15, "id": "93b4cf77", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:24.016580Z", "iopub.status.busy": "2026-07-31T12:20:24.015953Z", "iopub.status.idle": "2026-07-31T12:20:24.025025Z", "shell.execute_reply": "2026-07-31T12:20:24.024306Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Saved backtest returns to ../data/processed/backtest_returns.csv\n" ] } ], "source": [ "\"\"\"\n", "==================================\n", "Save backtest returns for risk decomposition\n", "==================================\n", "\"\"\"\n", "df_port[['long_net', 'short_net', 'ls_net', 'long_excess', 'ls_excess']].to_csv(\n", " '../data/processed/backtest_returns.csv'\n", ")\n", "\n", "print(\"Saved backtest returns to ../data/processed/backtest_returns.csv\")" ] }, { "cell_type": "markdown", "id": "8d50b216", "metadata": {}, "source": [ "## Robustness: Decile Size and Rebalance Timing\n", "\n", "The headline portfolio uses one specific set of choices: top 10%, monthly rebalance, equal-weight holdings. Before leaning on the alpha, we should vary the nearby choices that do not require new data.\n", "\n", "This grid varies:\n", "\n", "- **Decile cutoff:** 5%, 10%, 15%, 20%.\n", "- **Rebalance interval:** every 1, 2, or 3 months.\n", "\n", "For each combination, we recompute net returns, Sharpe, max drawdown, Fama-French alpha, ordinary OLS t-stat, HAC/Newey-West t-stat, and market beta.\n", "\n", "This still does not solve universe-construction robustness. The current-constituent universe remains a limitation. But it does answer whether the result depends on exactly one decile/rebalance setting." ] }, { "cell_type": "code", "execution_count": 16, "id": "970cd763", "metadata": { "execution": { "iopub.execute_input": "2026-07-31T12:20:24.027280Z", "iopub.status.busy": "2026-07-31T12:20:24.027058Z", "iopub.status.idle": "2026-07-31T12:20:27.934145Z", "shell.execute_reply": "2026-07-31T12:20:27.933511Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Long-only robustness across decile size and rebalance interval:\n", "\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
decilerebalance_every_monthsn_observationsann_returnsharpemax_drawdownalpha_annualizedalpha_t_olsalpha_t_hacmkt_betar_squared
05%12390.21961.0230-0.59370.07883.81364.05751.24600.8271
15%22390.19840.9023-0.64310.05472.64982.88791.28140.8355
25%32390.21651.0064-0.59460.07263.65773.74321.27240.8410
310%12390.19481.0036-0.57130.05973.97774.16561.18890.8885
410%22390.18270.9256-0.61370.04602.92782.92441.20770.8818
510%32390.18720.9568-0.59740.04893.38113.10631.21600.8979
615%12390.18250.9848-0.56640.04973.88163.89001.16500.9109
715%22390.18020.9652-0.56500.04733.75513.91901.16930.9150
815%32390.18080.9730-0.55300.04773.88943.89091.16820.9187
920%12390.17450.9718-0.55070.04583.93474.09601.13020.9217
1020%22390.17430.9658-0.54540.04484.05954.28991.13830.9302
1120%32390.17190.9553-0.53710.04223.80933.84341.13660.9294
\n", "
" ], "text/plain": [ " decile rebalance_every_months n_observations ann_return sharpe \\\n", "0 5% 1 239 0.2196 1.0230 \n", "1 5% 2 239 0.1984 0.9023 \n", "2 5% 3 239 0.2165 1.0064 \n", "3 10% 1 239 0.1948 1.0036 \n", "4 10% 2 239 0.1827 0.9256 \n", "5 10% 3 239 0.1872 0.9568 \n", "6 15% 1 239 0.1825 0.9848 \n", "7 15% 2 239 0.1802 0.9652 \n", "8 15% 3 239 0.1808 0.9730 \n", "9 20% 1 239 0.1745 0.9718 \n", "10 20% 2 239 0.1743 0.9658 \n", "11 20% 3 239 0.1719 0.9553 \n", "\n", " max_drawdown alpha_annualized alpha_t_ols alpha_t_hac mkt_beta \\\n", "0 -0.5937 0.0788 3.8136 4.0575 1.2460 \n", "1 -0.6431 0.0547 2.6498 2.8879 1.2814 \n", "2 -0.5946 0.0726 3.6577 3.7432 1.2724 \n", "3 -0.5713 0.0597 3.9777 4.1656 1.1889 \n", "4 -0.6137 0.0460 2.9278 2.9244 1.2077 \n", "5 -0.5974 0.0489 3.3811 3.1063 1.2160 \n", "6 -0.5664 0.0497 3.8816 3.8900 1.1650 \n", "7 -0.5650 0.0473 3.7551 3.9190 1.1693 \n", "8 -0.5530 0.0477 3.8894 3.8909 1.1682 \n", "9 -0.5507 0.0458 3.9347 4.0960 1.1302 \n", "10 -0.5454 0.0448 4.0595 4.2899 1.1383 \n", "11 -0.5371 0.0422 3.8093 3.8434 1.1366 \n", "\n", " r_squared \n", "0 0.8271 \n", "1 0.8355 \n", "2 0.8410 \n", "3 0.8885 \n", "4 0.8818 \n", "5 0.8979 \n", "6 0.9109 \n", "7 0.9150 \n", "8 0.9187 \n", "9 0.9217 \n", "10 0.9302 \n", "11 0.9294 " ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Headline setting from grid:\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
decilerebalance_every_monthsn_observationsann_returnsharpemax_drawdownalpha_annualizedalpha_t_olsalpha_t_hacmkt_betar_squared
310%12390.19481.0036-0.57130.05973.97774.16561.18890.8885
\n", "
" ], "text/plain": [ " decile rebalance_every_months n_observations ann_return sharpe \\\n", "3 10% 1 239 0.1948 1.0036 \n", "\n", " max_drawdown alpha_annualized alpha_t_ols alpha_t_hac mkt_beta \\\n", "3 -0.5713 0.0597 3.9777 4.1656 1.1889 \n", "\n", " r_squared \n", "3 0.8885 " ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "\n", "Interpretation: robust results should not require exactly one cutoff or one rebalance calendar. This grid still uses the same current-constituent universe, so universe robustness remains unresolved.\n" ] } ], "source": [ "\"\"\"\n", "==================================\n", "Robustness sweep: decile size × rebalance interval, with FF alpha\n", "==================================\n", "\"\"\"\n", "decile_grid = [0.05, 0.10, 0.15, 0.20]\n", "rebalance_grid = [1, 2, 3]\n", "tc_rate = TRANSACTION_COST_BPS / 10000\n", "\n", "\n", "def form_decile_portfolios_interval(signal_df, return_df, decile=0.1, rebalance_every=1):\n", " \"\"\"Long-only top-decile portfolio that rebalances every N months and holds between rebalances.\"\"\"\n", " common_dates = signal_df.index.intersection(return_df.index)\n", " common_tickers = signal_df.columns.intersection(return_df.columns)\n", " signal_df = signal_df.loc[common_dates, common_tickers]\n", " return_df = return_df.loc[common_dates, common_tickers]\n", "\n", " rows = []\n", " rebalance_dates = []\n", " current_holdings = None\n", " prev_w = None\n", "\n", " for i in range(len(common_dates) - 1):\n", " date = common_dates[i]\n", " next_date = common_dates[i + 1]\n", " should_rebalance = current_holdings is None or (i % rebalance_every == 0)\n", "\n", " if should_rebalance:\n", " scores = signal_df.loc[date].dropna()\n", " if len(scores) < 50:\n", " continue\n", " n_long = max(int(len(scores) * decile), 1)\n", " current_holdings = scores.sort_values(ascending=False).head(n_long).index.tolist()\n", " curr_w = _equal_weight_vector(current_holdings)\n", " if prev_w is None or prev_w.empty:\n", " turnover = curr_w.abs().sum()\n", " else:\n", " names = prev_w.index.union(curr_w.index)\n", " turnover = (curr_w.reindex(names, fill_value=0.0) - prev_w.reindex(names, fill_value=0.0)).abs().sum() / 2.0\n", " prev_w = curr_w\n", " else:\n", " turnover = 0.0\n", "\n", " next_rets = return_df.loc[next_date, current_holdings].dropna()\n", " if len(next_rets) == 0:\n", " continue\n", " long_ret = next_rets.mean()\n", " rows.append({\n", " 'long': long_ret,\n", " 'long_holdings': list(current_holdings),\n", " 'long_turnover': float(turnover),\n", " })\n", " rebalance_dates.append(next_date)\n", "\n", " out = pd.DataFrame(rows, index=pd.DatetimeIndex(rebalance_dates))\n", " out['long_net'] = out['long'] - out['long_turnover'] * tc_rate\n", " return out\n", "\n", "\n", "def ff_alpha_for_returns(return_series):\n", " \"\"\"FF alpha diagnostics for a long-only net return series.\"\"\"\n", " ret = return_series.rename('long_net').dropna().to_frame()\n", " ret.index = ret.index.to_period('M')\n", " ff_local = df_ff[['RF'] + FF_FACTOR_COLS].copy()\n", " ff_local.index = ff_local.index.to_period('M')\n", " reg = ret.join(ff_local, how='inner').dropna().astype(float)\n", " if len(reg) < 24:\n", " return pd.Series({\n", " 'alpha_annualized': np.nan,\n", " 'alpha_t_ols': np.nan,\n", " 'alpha_t_hac': np.nan,\n", " 'mkt_beta': np.nan,\n", " 'r_squared': np.nan,\n", " })\n", " y = reg['long_net'] - reg['RF']\n", " X = sm.add_constant(reg[FF_FACTOR_COLS], has_constant='add')\n", " model = sm.OLS(y, X).fit()\n", " model_hac = sm.OLS(y, X).fit(cov_type='HAC', cov_kwds={'maxlags': HAC_LAGS})\n", " return pd.Series({\n", " 'alpha_annualized': model.params['const'] * 12,\n", " 'alpha_t_ols': model.tvalues['const'],\n", " 'alpha_t_hac': model_hac.tvalues['const'],\n", " 'mkt_beta': model.params['Mkt-RF'],\n", " 'r_squared': model.rsquared,\n", " })\n", "\n", "robustness_rows = []\n", "for d in decile_grid:\n", " for rebalance_every in rebalance_grid:\n", " df_port_d = form_decile_portfolios_interval(\n", " df_momentum, df_returns, decile=d, rebalance_every=rebalance_every\n", " )\n", " m = performance_metrics(df_port_d['long_net'])\n", " alpha_diag = ff_alpha_for_returns(df_port_d['long_net'])\n", " robustness_rows.append({\n", " 'decile': f'{int(d*100)}%',\n", " 'rebalance_every_months': rebalance_every,\n", " 'n_observations': len(df_port_d),\n", " 'ann_return': m['ann_return'],\n", " 'sharpe': m['sharpe'],\n", " 'max_drawdown': m['max_drawdown'],\n", " **alpha_diag.to_dict(),\n", " })\n", "\n", "df_robustness = pd.DataFrame(robustness_rows)\n", "print(\"Long-only robustness across decile size and rebalance interval:\\n\")\n", "display(df_robustness.round(4))\n", "\n", "fig, axes = plt.subplots(1, 2, figsize=(13, 5))\n", "for rebalance_every in rebalance_grid:\n", " sub = df_robustness[df_robustness['rebalance_every_months'] == rebalance_every]\n", " axes[0].plot(sub['decile'], sub['sharpe'], marker='o', label=f'every {rebalance_every}m')\n", " axes[1].plot(sub['decile'], sub['alpha_t_hac'], marker='o', label=f'every {rebalance_every}m')\n", "\n", "axes[0].set_ylabel('Sharpe ratio')\n", "axes[0].set_title('Sharpe Across Robustness Grid')\n", "axes[0].grid(alpha=0.3)\n", "axes[1].axhline(y=2, color='coral', linestyle='--', linewidth=1, label='t = 2')\n", "axes[1].set_ylabel('HAC alpha t-stat')\n", "axes[1].set_title('FF Alpha Significance Across Grid')\n", "axes[1].grid(alpha=0.3)\n", "for ax in axes:\n", " ax.set_xlabel('Long cutoff')\n", " ax.legend(fontsize=8)\n", "\n", "plt.tight_layout()\n", "plt.savefig('../images/04_backtest/robustness_grid.png', dpi=150, bbox_inches='tight')\n", "plt.show()\n", "\n", "headline_row = df_robustness[\n", " (df_robustness['decile'] == '10%') &\n", " (df_robustness['rebalance_every_months'] == 1)\n", "]\n", "if not headline_row.empty:\n", " print(\"Headline setting from grid:\")\n", " display(headline_row.round(4))\n", "\n", "print(\n", " \"\\nInterpretation: robust results should not require exactly one cutoff or one rebalance calendar. \"\n", " \"This grid still uses the same current-constituent universe, so universe robustness remains unresolved.\"\n", ")" ] }, { "cell_type": "markdown", "id": "75addbfc", "metadata": {}, "source": [ "## Conclusion\n", "\n", "The main result is that a sector-neutralized momentum signal, traded long-only, produces a positive Fama-French alpha in this dataset. The result is more nuanced than the raw headline number:\n", "\n", "- **Long-only momentum:** Fama-French 4-factor alpha is about **5.97% annualized**. The ordinary OLS t-stat is **3.98**, and the HAC/Newey-West t-stat is **4.17**, so the result is not weakened by this simple autocorrelation/heteroskedasticity check.\n", "- **Market beta:** the long-only book has market beta around **1.19**, so much of the raw active return versus the equal-weight universe is market exposure. The FF alpha is the cleaner result because it controls for that beta.\n", "- **Walk-forward:** long-only Sharpe is positive in all four 5-year windows, but active return versus the equal-weight universe is positive in only three of four. The weak window is 2006-2011, which includes the 2008-09 momentum crash.\n", "- **Survivorship sensitivity:** the alpha survives **2%** annual synthetic drag under both flat and crash-concentrated assumptions. At **3%**, the flat drag is borderline, while the crash-concentrated drag loses significance. This is a breakeven stress test, not a true replacement for survivorship-free data.\n", "- **Robustness grid:** nearby decile and rebalance choices are checked with Sharpe and FF alpha. Alpha remains positive and HAC-significant across the tested 5-20% cutoffs and 1/2/3-month rebalance intervals, though universe-construction robustness is still unresolved.\n", "- **Long-short:** the long-short portfolio does not produce a significant alpha. In this universe, the useful part of the signal is concentrated on the long side; the short side adds noise.\n", "\n", "The next notebook decomposes the portfolio's risk, the quadratic form $w^\\top\\Sigma w$, into systematic and idiosyncratic components with PCA." ] } ], "metadata": { "kernelspec": { "display_name": ".venv", "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": "3.14.6" } }, "nbformat": 4, "nbformat_minor": 5 }