Portfolio intelligence / Regime Alpha

Know what carries
your risk.
And when it changes.

Go beyond the allocation table. Compare the drivers of portfolio risk across market regimes, with the evidence behind every number.

Explore portfolio drivers ↗See the setup

Built for allocators, advisers and investment teams.
Request access · Sign in

REGIME ALPHA / PORTFOLIO REVIEW

Synthetic mechanics demonstration

Same portfolio. Different risk drivers.

Fictional equity + bond sleeves · generated prices
50 / 50 initial capital · 3 Jan 2024–3 Feb 2025

Risk contributionOverall vs State 0

OverallState 0 · probability-weighted

Fictional equity has the largest signed variance share in State 0: 77.5%. Its share changes by -28.0 percentage points from overall. Shares also change when total portfolio variance changes; inspect absolute contributions alongside them.

Actual engine calculation on synthetic inputs. No real account or empirical investment performance is shown. Explore all regimes and exact values →

Read the evidence behind the line

Different signals.
A fuller picture.

Real observations. Declared units. Visible gaps. Explore the economic context alongside portfolio evidence.

The portfolio demo above is synthetic. A real-instrument portfolio example remains unavailable until verified adjusted-return inputs and publication provenance are attached.

PricesCPIAUCSL → 12-month change

LaborUNRATE → unemployment rate

PolicyFEDFUNDS → effective rate

This frozen public example uses revised observations retrieved on 14 September 2026. It is not a live feed or a point-in-time backtest. Missing October 2025 observations remain gaps.

Download source values Sources & calculation

Real economic history

FRED · Fed · BLS

Inflation and policy rates

January 2020–August 2026 · monthly · percent

CPI inflation · YoYEffective federal funds rate

Actual monthly US CPI inflation and federal funds rate observations, with missing observations retained as gaps.

CPI inflation · YoY: 3.35%
Effective policy rate: 3.63%
Unemployment: 4.1%

August 2026 observations · retrieved 14 September 2026

Sources: BLS CPI via FRED · Federal Reserve rate via FRED. Macroeconomic context; these are not portfolio risk contributions.

View unemployment history

US unemployment rate from actual BLS observations.

Source: BLS unemployment via FRED. Percent, seasonally adjusted.

Inside the workflow

Your inputs. An explicit Build.
Evidence you can inspect.

Illustrated guides based on the current setup controls. Example inputs, not account screenshots or recommended settings.

Portfolio allocation guide: example SPY 60 percent and AGG 40 percent, with buy-and-hold accounting.

01 / Portfolio

Start with what you hold.

Enter tickers and capital weights. Confirm the accounting policy and supported price history.

02 / Features

Know what each input means.

Search the variable catalog. Check source, units, transform and available history before selecting features.

Model configuration guide: choose model type, number of regimes and training dates, then press Build Model.

03 / Model & Build

Build once. Review visually.

Set the model and training cutoff, then Build. Review portfolio history, drawdown, risk shares and regime probabilities.

Edits remain drafts. Completed results keep their original inputs until the next successful Build. Inspect the separately labeled synthetic mechanics example →

Methods & supported scope

Trace the chart
back to the evidence.

Long-only USD portfolios, supported adjusted-return history and explicit cash treatment. Buy-and-hold or scheduled rebalancing. Charts retain the dates, policy and sources of the saved analysis.

Signed regression exposures, covariance risk and probability-weighted historical structure support review. Exact tables remain available for inspection and export.

Keep interpretation grounded.

Current-holdings history is not your account performance record.
Historical associations are not causal explanations or forward stress forecasts.
Regime membership probability is not forecast calibration.
Missing data stays unavailable. A readiness pass does not validate an investment strategy.

Inspect the numerical sample's methods

Built by Joe Bunster

A research-led product.

Applied mathematics and quantitative modeling, brought to portfolio review.

Joe's professional background

Review your portfolio.

Access is approval-based. Email Joe about your use case to confirm availability and the account to use.

Request access Sign in to your approved account

Know what carriesyour risk.And when it changes.