{

“cells”: [

{

“cell_type”: “markdown”, “id”: “intro-md”, “metadata”: {}, “source”: [

“# ETF Multi-Asset Walkthroughn”, “n”, “End-to-end portfolio construction pipeline using py-vAllocation:n”, “n”, “1. Load and prepare weekly log-returns for 4 ETFs (SPY, TLT, GLD, DBC).n”, “2. Estimate moments with James-Stein shrinkage + OAS covariance.n”, “3. Project to a 1-year investment horizon.n”, “4. Build Mean-Variance, CVaR, and Risk Parity frontiers.n”, “5. Select portfolios at comparable risk and blend via exposure stacking.n”, “6. Stress-test the blended portfolio across scenarios and horizons.n”, “7. Discrete allocation into integer share counts.”

]

}, {

“cell_type”: “code”, “execution_count”: 1, “id”: “setup”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:19.012390Z”, “iopub.status.busy”: “2026-03-30T19:38:19.012275Z”, “iopub.status.idle”: “2026-03-30T19:38:19.702680Z”, “shell.execute_reply”: “2026-03-30T19:38:19.702412Z”

}

}, “outputs”: [], “source”: [

“%matplotlib inlinen”, “import numpy as npn”, “import pandas as pdn”, “import matplotlib.pyplot as pltn”, “import seaborn as snsn”, “from pathlib import Pathn”, “n”, “np.random.seed(42)n”, “sns.set_style(‘whitegrid’)”

]

}, {

“cell_type”: “markdown”, “id”: “load-data-md”, “metadata”: {}, “source”: [

“## Load ETF prices and compute weekly log-returns”

]

}, {

“cell_type”: “code”, “execution_count”: 2, “id”: “load-data”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:19.703939Z”, “iopub.status.busy”: “2026-03-30T19:38:19.703834Z”, “iopub.status.idle”: “2026-03-30T19:38:19.716974Z”, “shell.execute_reply”: “2026-03-30T19:38:19.716774Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“Assets: [‘SPY’, ‘TLT’, ‘GLD’, ‘DBC’]n”, “Weeks : 1006 (2006-02-17 to 2025-05-23)n”

]

}

], “source”: [

“_candidates = [n”, “ Path(‘examples/ETF_prices.csv’),n”, “ Path(‘../examples/ETF_prices.csv’),n”, “ Path(‘../../examples/ETF_prices.csv’),n”, “ Path(‘../../../examples/ETF_prices.csv’),n”, “]n”, “_csv = next((p for p in _candidates if p.exists()), None)n”, “if _csv is None:n”, “ raise FileNotFoundError(‘ETF_prices.csv not found’)n”, “n”, “prices = pd.read_csv(_csv, index_col=’Date’, parse_dates=True)n”, “prices = prices[[‘SPY’, ‘TLT’, ‘GLD’, ‘DBC’]].dropna().ffill()n”, “weekly = prices.resample(‘W-FRI’).last().dropna()n”, “log_returns = np.log(weekly).diff().dropna()n”, “n”, “print(f’Assets: {list(log_returns.columns)}’)n”, “print(f’Weeks : {len(log_returns)} ({log_returns.index[0].date()} to {log_returns.index[-1].date()})’)”

]

}, {

“cell_type”: “code”, “execution_count”: 3, “id”: “train-test-split”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:19.717999Z”, “iopub.status.busy”: “2026-03-30T19:38:19.717925Z”, “iopub.status.idle”: “2026-03-30T19:38:19.719946Z”, “shell.execute_reply”: “2026-03-30T19:38:19.719748Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“In-sample : 804 weeksn”, “Out-of-sample: 202 weeksn”

]

}

], “source”: [

“split = int(len(log_returns) * 0.8)n”, “log_ret_is = log_returns.iloc[:split] # in-samplen”, “log_ret_oos = log_returns.iloc[split:] # out-of-samplen”, “n”, “simple_ret_oos = np.exp(log_ret_oos) - 1 # simple returns for performancen”, “n”, “print(f’In-sample : {len(log_ret_is)} weeks’)n”, “print(f’Out-of-sample: {len(log_ret_oos)} weeks’)”

]

}, {

“cell_type”: “markdown”, “id”: “step1-md”, “metadata”: {}, “source”: [

“## Step 1 – Estimate moments (P2)n”, “n”, “James-Stein shrinkage for the mean vector and Oracle Approximating Shrinkage (OAS)n”, “for the covariance matrix.”

]

}, {

“cell_type”: “code”, “execution_count”: 4, “id”: “step1-estimate”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:19.720856Z”, “iopub.status.busy”: “2026-03-30T19:38:19.720782Z”, “iopub.status.idle”: “2026-03-30T19:38:19.739206Z”, “shell.execute_reply”: “2026-03-30T19:38:19.739012Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“Weekly expected log-returns (bps):n”, “SPY 16.58n”, “TLT 11.68n”, “GLD 13.06n”, “DBC 2.65n”, “Name: mu, dtype: float64n”

]

}

], “source”: [

“from pyvallocation.moments import estimate_momentsn”, “n”, “mu, cov = estimate_moments(n”, “ log_ret_is,n”, “ mean_estimator=’james_stein’,n”, “ cov_estimator=’oas’,n”, “)n”, “n”, “print(‘Weekly expected log-returns (bps):’)n”, “print((mu * 1e4).round(2))”

]

}, {

“cell_type”: “code”, “execution_count”: 5, “id”: “step1-heatmap”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:19.740196Z”, “iopub.status.busy”: “2026-03-30T19:38:19.740141Z”, “iopub.status.idle”: “2026-03-30T19:38:19.793228Z”, “shell.execute_reply”: “2026-03-30T19:38:19.793025Z”

}

}, “outputs”: [

{

“data”: {

“image/png”: “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”, “text/plain”: [

“<Figure size 500x400 with 2 Axes>”

]

}, “metadata”: {}, “output_type”: “display_data”

}

], “source”: [

“corr = pd.DataFrame(n”, “ cov.values / np.outer(np.sqrt(np.diag(cov.values)), np.sqrt(np.diag(cov.values))),n”, “ index=cov.index, columns=cov.columns,n”, “)n”, “fig, ax = plt.subplots(figsize=(5, 4))n”, “sns.heatmap(corr, annot=True, fmt=’.2f’, cmap=’RdBu_r’, center=0, ax=ax)n”, “ax.set_title(‘OAS Correlation Matrix’)n”, “plt.tight_layout()n”, “plt.show()”

]

}, {

“cell_type”: “markdown”, “id”: “step2-md”, “metadata”: {}, “source”: [

“## Step 2 – Project to 1-year horizon (P3 + P4)n”, “n”, “Scale the weekly moments to an annual horizon and convert from log to simple returns.”

]

}, {

“cell_type”: “code”, “execution_count”: 6, “id”: “step2-project”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:19.794287Z”, “iopub.status.busy”: “2026-03-30T19:38:19.794227Z”, “iopub.status.idle”: “2026-03-30T19:38:19.798159Z”, “shell.execute_reply”: “2026-03-30T19:38:19.797956Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“ E[r] (%) Vol (%)n”, “SPY 10.91 20.79n”, “TLT 7.28 14.87n”, “GLD 8.73 19.48n”, “DBC 3.27 19.98n”

]

}

], “source”: [

“from pyvallocation.utils.projection import project_mean_covariance, log2simplen”, “n”, “ANNUALIZATION = 52n”, “mu_h, cov_h = project_mean_covariance(mu, cov, annualization_factor=ANNUALIZATION)n”, “mu_s, cov_s = log2simple(mu_h, cov_h)n”, “n”, “annual = pd.DataFrame({n”, “ ‘E[r] (%)’: (mu_s * 100).round(2),n”, “ ‘Vol (%)’: (np.sqrt(np.diag(cov_s.values)) * 100).round(2),n”, “})n”, “print(annual)”

]

}, {

“cell_type”: “markdown”, “id”: “step3-md”, “metadata”: {}, “source”: [

“## Step 3 – Build multiple frontiersn”, “n”, “Construct Mean-Variance, CVaR, and Relaxed Risk Parity frontiers on the annual simple-return moments.”

]

}, {

“cell_type”: “code”, “execution_count”: 7, “id”: “step3-frontiers”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:19.799450Z”, “iopub.status.busy”: “2026-03-30T19:38:19.799389Z”, “iopub.status.idle”: “2026-03-30T19:38:26.780025Z”, “shell.execute_reply”: “2026-03-30T19:38:26.779573Z”

}

}, “outputs”: [

{

“name”: “stderr”, “output_type”: “stream”, “text”: [

“Scenario simulation is non-reproducible (no seed). Pass seed= for reproducibility.n”

]

}, {

“name”: “stderr”, “output_type”: “stream”, “text”: [

“/Users/giuliocantadori/dev/Py-vAllocation/pyvallocation/portfolioapi.py:987: RuntimeWarning: divide by zero encountered in matmuln”, “ scenarios = rng.multivariate_normal(self.dist.mu, self.dist.cov, n_simulations)n”, “/Users/giuliocantadori/dev/Py-vAllocation/pyvallocation/portfolioapi.py:987: RuntimeWarning: overflow encountered in matmuln”, “ scenarios = rng.multivariate_normal(self.dist.mu, self.dist.cov, n_simulations)n”, “/Users/giuliocantadori/dev/Py-vAllocation/pyvallocation/portfolioapi.py:987: RuntimeWarning: invalid value encountered in matmuln”, “ scenarios = rng.multivariate_normal(self.dist.mu, self.dist.cov, n_simulations)n”

]

}

], “source”: [

“from pyvallocation import PortfolioWrapper, plot_frontiersn”, “n”, “wrapper_mv = PortfolioWrapper.from_moments(mu_s, cov_s)n”, “frontier_mv = wrapper_mv.variance_frontier(num_portfolios=10)n”, “n”, “wrapper_cvar = PortfolioWrapper.from_moments(mu_s, cov_s)n”, “frontier_cvar = wrapper_cvar.cvar_frontier(num_portfolios=10, alpha=0.05)n”, “n”, “wrapper_rp = PortfolioWrapper.from_moments(mu_s, cov_s)n”, “frontier_rp = wrapper_rp.relaxed_risk_parity_frontier(n”, “ num_portfolios=10, lambda_reg=0.0,n”, “)”

]

}, {

“cell_type”: “code”, “execution_count”: 8, “id”: “step3-plot”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:26.781988Z”, “iopub.status.busy”: “2026-03-30T19:38:26.781850Z”, “iopub.status.idle”: “2026-03-30T19:38:26.832161Z”, “shell.execute_reply”: “2026-03-30T19:38:26.831942Z”

}

}, “outputs”: [

{

“data”: {

“image/png”: “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”, “text/plain”: [

“<Figure size 800x500 with 1 Axes>”

]

}, “metadata”: {}, “output_type”: “display_data”

}

], “source”: [

“fig, ax = plt.subplots(figsize=(8, 5))n”, “for frontier, label, marker in [n”, “ (frontier_mv, ‘Mean-Variance’, ‘o’),n”, “ (frontier_cvar, ‘CVaR (alpha=5%)’, ‘s’),n”, “ (frontier_rp, ‘Risk Parity’, ‘^’),n”, “]:n”, “ ax.plot(frontier.risks * 100, frontier.returns * 100,n”, “ marker=marker, label=label, markersize=5)n”, “ax.set_xlabel(f’Risk (%) [{frontier_mv.risk_measure}]’)n”, “ax.set_ylabel(‘Expected Return (%)’)n”, “ax.set_title(‘Efficient Frontiers Comparison’)n”, “ax.legend()n”, “plt.tight_layout()n”, “plt.show()”

]

}, {

“cell_type”: “markdown”, “id”: “step4-md”, “metadata”: {}, “source”: [

“## Step 4 – Select portfolios at comparable riskn”, “n”, “Pick the 50th risk-percentile portfolio from each frontier so they sit at roughly the same risk level.”

]

}, {

“cell_type”: “code”, “execution_count”: 9, “id”: “step4-select”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:26.833261Z”, “iopub.status.busy”: “2026-03-30T19:38:26.833186Z”, “iopub.status.idle”: “2026-03-30T19:38:26.837105Z”, “shell.execute_reply”: “2026-03-30T19:38:26.836908Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“ MV CVaR RPn”, “SPY 0.4113 0.5456 0.4765n”, “TLT 0.4258 0.2809 0.3398n”, “GLD 0.1629 0.1734 0.1067n”, “DBC 0.0000 0.0000 0.0770n”

]

}

], “source”: [

“from pyvallocation.ensembles import risk_percentile_selectionsn”, “n”, “sels = risk_percentile_selections(n”, “ [frontier_mv, frontier_cvar, frontier_rp], percentile=0.50,n”, “)n”, “n”, “w_mv, r_mv, s_mv = frontier_mv.at_percentile(0.50)n”, “w_cvar, r_cvar, s_cvar = frontier_cvar.at_percentile(0.50)n”, “w_rp, r_rp, s_rp = frontier_rp.at_percentile(0.50)n”, “n”, “portfolios = {‘MV’: w_mv, ‘CVaR’: w_cvar, ‘RP’: w_rp}n”, “n”, “comparison = pd.DataFrame(portfolios).round(4)n”, “print(comparison)”

]

}, {

“cell_type”: “code”, “execution_count”: 10, “id”: “step4-bar”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:26.838178Z”, “iopub.status.busy”: “2026-03-30T19:38:26.838098Z”, “iopub.status.idle”: “2026-03-30T19:38:26.879522Z”, “shell.execute_reply”: “2026-03-30T19:38:26.879305Z”

}

}, “outputs”: [

{

“data”: {

“image/png”: “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”, “text/plain”: [

“<Figure size 700x400 with 1 Axes>”

]

}, “metadata”: {}, “output_type”: “display_data”

}

], “source”: [

“comparison.plot.bar(figsize=(7, 4), rot=0)n”, “plt.ylabel(‘Weight’)n”, “plt.title(‘Selected Portfolio Weights (50th risk percentile)’)n”, “plt.tight_layout()n”, “plt.show()”

]

}, {

“cell_type”: “markdown”, “id”: “step5-md”, “metadata”: {}, “source”: [

“## Step 5 – Ensemble blendingn”, “n”, “Combine the three selected portfolios via exposure stacking and simple averaging.”

]

}, {

“cell_type”: “code”, “execution_count”: 11, “id”: “step5-ensemble”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:26.880592Z”, “iopub.status.busy”: “2026-03-30T19:38:26.880530Z”, “iopub.status.idle”: “2026-03-30T19:38:26.888227Z”, “shell.execute_reply”: “2026-03-30T19:38:26.888018Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“ Stacked Averagen”, “SPY 0.4705 0.4778n”, “TLT 0.3603 0.3489n”, “GLD 0.1611 0.1477n”, “DBC 0.0081 0.0257n”

]

}

], “source”: [

“from pyvallocation.ensembles import exposure_stacking, average_exposuresn”, “n”, “samples = pd.DataFrame(portfolios)n”, “w_stacked = exposure_stacking(samples, L=2)n”, “w_avg = average_exposures(samples)n”, “n”, “blend = pd.DataFrame({‘Stacked’: w_stacked, ‘Average’: w_avg}).round(4)n”, “print(blend)n”, “n”, “# Use stacked weights for downstream analysisn”, “weights = np.asarray(w_stacked, dtype=float).ravel()”

]

}, {

“cell_type”: “markdown”, “id”: “step6-md”, “metadata”: {}, “source”: [

“## Step 6 – Stress testingn”, “n”, “Evaluate the stacked portfolio under nominal, decay, kernel-focus, and shock scenarios,n”, “then project across multiple horizons.”

]

}, {

“cell_type”: “markdown”, “id”: “step6a-md”, “metadata”: {}, “source”: [

“### 6a. Nominal performance (out-of-sample weekly returns)”

]

}, {

“cell_type”: “code”, “execution_count”: 12, “id”: “step6a”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:26.889320Z”, “iopub.status.busy”: “2026-03-30T19:38:26.889260Z”, “iopub.status.idle”: “2026-03-30T19:38:26.891671Z”, “shell.execute_reply”: “2026-03-30T19:38:26.891471Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“=== Nominal Performance (OOS weekly) ===n”, “mean 0.000769n”, “stdev 0.015344n”, “VaR95 0.023400n”, “CVaR95 0.028934n”, “ENS 202.000000n”, “dtype: float64n”

]

}

], “source”: [

“from pyvallocation.utils.performance import performance_reportn”, “n”, “report_nom = performance_report(weights, simple_ret_oos.values, confidence=0.95)n”, “print(‘=== Nominal Performance (OOS weekly) ===’)n”, “print(report_nom.round(6))”

]

}, {

“cell_type”: “markdown”, “id”: “step6b-md”, “metadata”: {}, “source”: [

“### 6b. Exponential-decay stress (half-life = 30 weeks)”

]

}, {

“cell_type”: “code”, “execution_count”: 13, “id”: “step6b”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:26.892761Z”, “iopub.status.busy”: “2026-03-30T19:38:26.892707Z”, “iopub.status.idle”: “2026-03-30T19:38:26.895868Z”, “shell.execute_reply”: “2026-03-30T19:38:26.895669Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“=== Exponential Decay Stress (HL=30) ===n”, “ return_nom stdev_nom VaR95_nom CVaR95_nom ENS_nom \n”, “portfolio_0 0.000769 0.015344 0.0234 0.028934 202.0 n”, “n”, “ return_stress stdev_stress VaR95_stress CVaR95_stress \n”, “portfolio_0 0.001794 0.014325 0.02033 0.028524 n”, “n”, “ ENS_stress KL_q_p n”, “portfolio_0 111.498157 0.59426 n”

]

}

], “source”: [

“from pyvallocation.stress import exp_decay_stressn”, “n”, “report_decay = exp_decay_stress(weights, simple_ret_oos.values, half_life=30)n”, “print(‘=== Exponential Decay Stress (HL=30) ===’)n”, “print(report_decay.round(6))”

]

}, {

“cell_type”: “markdown”, “id”: “step6c-md”, “metadata”: {}, “source”: [

“### 6c. Kernel-focus stress targeting the worst volatility regime”

]

}, {

“cell_type”: “code”, “execution_count”: 14, “id”: “step6c”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:26.896874Z”, “iopub.status.busy”: “2026-03-30T19:38:26.896820Z”, “iopub.status.idle”: “2026-03-30T19:38:26.901583Z”, “shell.execute_reply”: “2026-03-30T19:38:26.901410Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“=== Kernel Focus Stress (high-vol regime) ===n”, “ return_nom stdev_nom VaR95_nom CVaR95_nom ENS_nom \n”, “portfolio_0 0.000769 0.015344 0.0234 0.028934 202.0 n”, “n”, “ return_stress stdev_stress VaR95_stress CVaR95_stress \n”, “portfolio_0 -0.009651 0.013562 0.0234 0.02553 n”, “n”, “ ENS_stress KL_q_p n”, “portfolio_0 11.952093 2.827361 n”

]

}

], “source”: [

“from pyvallocation.stress import kernel_focus_stressn”, “n”, “vol_proxy = simple_ret_oos.std(axis=1).rolling(12).mean().bfill()n”, “n”, “report_kernel = kernel_focus_stress(n”, “ weights, simple_ret_oos.values,n”, “ focus_series=vol_proxy.values,n”, “ target=vol_proxy.values.max(),n”, “)n”, “print(‘=== Kernel Focus Stress (high-vol regime) ===’)n”, “print(report_kernel.round(6))”

]

}, {

“cell_type”: “markdown”, “id”: “step6d-md”, “metadata”: {}, “source”: [

“### 6d. Linear shock – scale up returns by 1.5x with a negative mean shift”

]

}, {

“cell_type”: “code”, “execution_count”: 15, “id”: “step6d”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:26.902504Z”, “iopub.status.busy”: “2026-03-30T19:38:26.902448Z”, “iopub.status.idle”: “2026-03-30T19:38:26.905334Z”, “shell.execute_reply”: “2026-03-30T19:38:26.905176Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“=== Linear Shock (1.5x scale, -30 bps shift) ===n”, “ return_nom stdev_nom VaR95_nom CVaR95_nom ENS_nom \n”, “portfolio_0 0.000769 0.015344 0.0234 0.028934 202.0 n”, “n”, “ return_stress stdev_stress VaR95_stress CVaR95_stress \n”, “portfolio_0 -0.001846 0.023017 0.0381 0.046401 n”, “n”, “ ENS_stress n”, “portfolio_0 202.0 n”

]

}

], “source”: [

“from pyvallocation.stress import stress_test, linear_mapn”, “n”, “shock = linear_map(scale=1.5, mean_shift=np.full(4, -0.003))n”, “report_shock = stress_test(weights, simple_ret_oos.values, transform=shock)n”, “print(‘=== Linear Shock (1.5x scale, -30 bps shift) ===’)n”, “print(report_shock.round(6))”

]

}, {

“cell_type”: “markdown”, “id”: “step6e-md”, “metadata”: {}, “source”: [

“### 6e. Multi-horizon projectionn”, “n”, “Bootstrap in-sample log-returns to 1-month, 3-month, and 6-month horizons,n”, “reprice via reprice_exp, and report performance.”

]

}, {

“cell_type”: “code”, “execution_count”: 16, “id”: “step6e”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:26.906364Z”, “iopub.status.busy”: “2026-03-30T19:38:26.906291Z”, “iopub.status.idle”: “2026-03-30T19:38:26.927007Z”, “shell.execute_reply”: “2026-03-30T19:38:26.926768Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“ mean stdev VaR95 CVaR95 ENSn”, “horizon_weeks n”, “4 0.0069 0.0260 0.0340 0.0555 5000.0n”, “13 0.0238 0.0477 0.0562 0.0823 5000.0n”, “26 0.0488 0.0688 0.0646 0.0941 5000.0n”

]

}

], “source”: [

“from pyvallocation.utils.projection import project_scenarios, reprice_expn”, “n”, “rows = []n”, “for hw in [4, 13, 26]:n”, “ scen = project_scenarios(n”, “ log_ret_is, investment_horizon=hw,n”, “ n_simulations=5000, reprice=reprice_exp,n”, “ )n”, “ rpt = performance_report(weights, scen.values, confidence=0.95)n”, “ rows.append({‘horizon_weeks’: hw, **rpt.to_dict()})n”, “n”, “horizon_table = pd.DataFrame(rows).set_index(‘horizon_weeks’)n”, “print(horizon_table.round(4))”

]

}, {

“cell_type”: “markdown”, “id”: “step6f-md”, “metadata”: {}, “source”: [

“### 6f. Path visualisation – return distributions by horizon and portfolio”

]

}, {

“cell_type”: “code”, “execution_count”: 17, “id”: “step6f”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:26.928136Z”, “iopub.status.busy”: “2026-03-30T19:38:26.928066Z”, “iopub.status.idle”: “2026-03-30T19:38:27.250484Z”, “shell.execute_reply”: “2026-03-30T19:38:27.250259Z”

}

}, “outputs”: [

{

“name”: “stderr”, “output_type”: “stream”, “text”: [

“/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: divide by zero encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: overflow encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: invalid value encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: divide by zero encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: overflow encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: invalid value encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: divide by zero encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: overflow encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: invalid value encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: divide by zero encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: overflow encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: invalid value encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: divide by zero encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: overflow encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: invalid value encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: divide by zero encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: overflow encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: invalid value encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: divide by zero encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: overflow encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: invalid value encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: divide by zero encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: overflow encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: invalid value encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: divide by zero encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: overflow encountered in matmuln”, “ pnl = scen.values @ w_arrn”, “/var/folders/kd/ch5dsx357yg2zrg8f0kr0nr00000gn/T/ipykernel_72175/3603837998.py:11: RuntimeWarning: invalid value encountered in matmuln”, “ pnl = scen.values @ w_arrn”

]

}, {

“data”: {

“image/png”: “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”, “text/plain”: [

“<Figure size 1500x400 with 3 Axes>”

]

}, “metadata”: {}, “output_type”: “display_data”

}

], “source”: [

“fig, axes = plt.subplots(1, 3, figsize=(15, 4))n”, “horizon_colors = {4: ‘steelblue’, 13: ‘darkorange’, 26: ‘firebrick’}n”, “n”, “for ax, (name, w_s) in zip(axes, portfolios.items()):n”, “ w_arr = np.asarray(w_s, dtype=float).ravel()n”, “ for hw, color in horizon_colors.items():n”, “ scen = project_scenarios(n”, “ log_ret_is, investment_horizon=hw,n”, “ n_simulations=5000, reprice=reprice_exp,n”, “ )n”, “ pnl = scen.values @ w_arrn”, “ ax.hist(pnl, bins=50, alpha=0.4, color=color, label=f’{hw}w’)n”, “ ax.set_title(name)n”, “ ax.legend()n”, “n”, “fig.suptitle(‘Portfolio Return Distribution by Horizon’)n”, “plt.tight_layout()n”, “plt.show()”

]

}, {

“cell_type”: “markdown”, “id”: “step6-summary-md”, “metadata”: {}, “source”: [

“### Stress summary table”

]

}, {

“cell_type”: “code”, “execution_count”: 18, “id”: “step6-summary”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:27.251603Z”, “iopub.status.busy”: “2026-03-30T19:38:27.251531Z”, “iopub.status.idle”: “2026-03-30T19:38:27.254647Z”, “shell.execute_reply”: “2026-03-30T19:38:27.254461Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“ Nominal (weekly) Exp-Decay (HL=30) Kernel Focus Linear Shockn”, “CVaR95 0.028934 NaN NaN NaNn”, “CVaR95_nom NaN 0.028934 0.028934 0.028934n”, “CVaR95_stress NaN 0.028524 0.025530 0.046401n”, “ENS 202.000000 NaN NaN NaNn”, “ENS_nom NaN 202.000000 202.000000 202.000000n”, “ENS_stress NaN 111.498157 11.952093 202.000000n”, “KL_q_p NaN 0.594260 2.827361 NaNn”, “VaR95 0.023400 NaN NaN NaNn”, “VaR95_nom NaN 0.023400 0.023400 0.023400n”, “VaR95_stress NaN 0.020330 0.023400 0.038100n”, “mean 0.000769 NaN NaN NaNn”, “return_nom NaN 0.000769 0.000769 0.000769n”, “return_stress NaN 0.001794 -0.009651 -0.001846n”, “stdev 0.015344 NaN NaN NaNn”, “stdev_nom NaN 0.015344 0.015344 0.015344n”, “stdev_stress NaN 0.014325 0.013562 0.023017n”

]

}

], “source”: [

“summary = pd.DataFrame({n”, “ ‘Nominal (weekly)’: report_nom,n”, “ ‘Exp-Decay (HL=30)’: report_decay.iloc[0],n”, “ ‘Kernel Focus’: report_kernel.iloc[0],n”, “ ‘Linear Shock’: report_shock.iloc[0],n”, “})n”, “print(summary.round(6))”

]

}, {

“cell_type”: “markdown”, “id”: “step7-md”, “metadata”: {}, “source”: [

“## Step 7 – Discrete allocationn”, “n”, “Convert the stacked continuous weights into integer share counts given a $100,000 portfolio,n”, “using the latest prices in the dataset.”

]

}, {

“cell_type”: “code”, “execution_count”: 19, “id”: “step7-discrete”, “metadata”: {

“execution”: {

“iopub.execute_input”: “2026-03-30T19:38:27.255649Z”, “iopub.status.busy”: “2026-03-30T19:38:27.255594Z”, “iopub.status.idle”: “2026-03-30T19:38:27.260370Z”, “shell.execute_reply”: “2026-03-30T19:38:27.260179Z”

}

}, “outputs”: [

{

“name”: “stdout”, “output_type”: “stream”, “text”: [

“Shares:n”, “ SPY: 81 shares @ $579.11 = $46,907.91n”, “ TLT: 427 shares @ $84.55 = $36,102.85n”, “ GLD: 52 shares @ $309.75 = $16,107.00n”, “ DBC: 38 shares @ $21.41 = $813.58n”, “n”, “Leftover cash: $68.66n”, “Tracking error (L1): 0.0021n”

]

}

], “source”: [

“from pyvallocation.discrete_allocation import discretize_weightsn”, “n”, “latest_prices = prices.iloc[-1]n”, “total_value = 100_000.0n”, “n”, “w_series = pd.Series(weights, index=log_returns.columns, name=’Stacked’)n”, “result = discretize_weights(w_series, latest_prices, total_value)n”, “n”, “print(‘Shares:’)n”, “for asset, count in result.shares.items():n”, “ print(f’ {asset}: {count} shares @ ${latest_prices[asset]:.2f} = ${count * latest_prices[asset]:,.2f}’)n”, “print(f’\nLeftover cash: ${result.leftover_cash:,.2f}’)n”, “print(f’Tracking error (L1): {result.tracking_error:.4f}’)”

]

}

], “metadata”: {

“kernelspec”: {

“display_name”: “Python 3 (ipykernel)”, “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”: “3.12.9”

}

}, “nbformat”: 4, “nbformat_minor”: 5

}