How the Risk Index Is Computed

The Risk Index is a total risk metric: it measures your portfolio's overall volatility (standard deviation of returns). It is not a partial or systematic-only measure—it captures total variance.

Formula

Risk Index = (Portfolio annualized volatility / SPY annualized volatility) × 100

Where:

  • Portfolio volatility = standard deviation of your portfolio's daily returns, annualized
  • SPY volatility = standard deviation of SPY's daily returns over the same period, annualized

Scale

| Value | Meaning | |-------|---------| | 0 | Cash-like (no volatility) | | 100 | Same total risk as the S&P 500 (SPY) | | > 100 | Higher volatility than the broad market | | < 100 | Lower volatility than the broad market |

Variance Decomposition

The total portfolio variance is decomposed into four components (market, sector, subsector, idiosyncratic). Your dashboard shows the percentage of risk attributable to each:

  • Market — Variance explained by broad market moves (SPY)
  • Sector — Variance from sector tilts (e.g., tech vs. financials)
  • Subsector — Variance from industry/subindustry tilts
  • Idiosyncratic — Stock-specific variance (the sub-sector residual, a proxy for pure alpha)

These proportions sum to 100% and show where your risk comes from; the Risk Index score itself uses total variance.

Data and Scope

  • Returns-based: Uses actual historical portfolio-weighted returns from the ERM3 model, not parametric approximations.
  • Window: Last 252 trading days (≈1 year).
  • Holdings: Only positions whose tickers are in our ERM3 universe contribute. Your dashboard shows "X/Y holdings analyzed"—only the X matching holdings drive the index and charts.
  • SPY normalization: If SPY return data is unavailable, we use an industry-standard 16% annualized volatility assumption.

Implementation: portfolio-analytics-tab.tsx (returns-based variance decomposition).