Ulcer Index
The Ulcer Index measures the depth and duration of drawdowns in a portfolio, quantifying the "pain" experienced by investors through sustained periods of decline. Unlike standard deviation, it only penalizes downside movements and captures the compounding effect of prolonged drawdowns.
Overview
The Ulcer Index was developed by Peter Martin and Byron McCann in 1987 as part of their work on mutual fund performance evaluation. The name reflects the idea that sustained drawdowns cause investors "ulcers" -- the measure captures the anxiety and stress associated with watching a portfolio decline over extended periods.
The Ulcer Index improves upon Maximum Drawdown by considering not just the worst single drawdown but the entire drawdown history. It is calculated as the root mean square of percentage drawdowns from the running peak, giving greater weight to larger and longer-lasting drawdowns. A portfolio that experiences a single sharp decline followed by a quick recovery will have a lower Ulcer Index than one that suffers prolonged, moderate losses.
The companion metric, the Ulcer Performance Index (UPI), also known as Martin Ratio, adjusts returns for drawdown risk, providing a risk-adjusted performance measure that is more relevant to investor experience than the Sharpe Ratio for long-only equity strategies.
Mathematical Formulation
Percentage Drawdown
At each observation point , the percentage drawdown from the running peak is:
where is the portfolio value at time and is the highest portfolio value observed up to and including time . Note that always; it equals zero when the portfolio is at a new peak and is negative otherwise.
Ulcer Index
The Ulcer Index is the root mean square of the percentage drawdowns over observations:
Because the drawdowns are squared before averaging, the Ulcer Index gives disproportionately greater weight to large drawdowns. A portfolio that is always at its peak has . Higher values indicate deeper and more prolonged drawdowns. Typical values for equity indices range from 5% to 15%, while well-managed funds may achieve values below 5%.
Ulcer Performance Index (UPI / Martin Ratio)
The UPI adjusts excess returns for drawdown risk, analogous to how the Sharpe Ratio adjusts for volatility:
where is the portfolio return, is the risk-free rate, and is the Ulcer Index. A higher UPI indicates better risk-adjusted performance, where risk is defined in terms of drawdown severity rather than overall volatility.
Comparison with Standard Deviation
The Ulcer Index differs from standard deviation in several important ways:
Standard Deviation
- Penalizes both upside and downside deviations equally
- Based on deviations from the mean return
- Independent of the order of returns
- Symmetric risk measure
Ulcer Index
- Penalizes only drawdowns (declines from peak)
- Based on deviations from the running maximum
- Path-dependent -- order of returns matters
- Asymmetric, downside-only risk measure
Worked Example
Consider a portfolio observed over 5 periods with values: 100, 108, 102, 95, 105.
| Period | Value | Running Max | ||
|---|---|---|---|---|
| 1 | 100 | 100 | 0.00% | 0.0000 |
| 2 | 108 | 108 | 0.00% | 0.0000 |
| 3 | 102 | 108 | -5.56% | 30.8642 |
| 4 | 95 | 108 | -12.04% | 144.9216 |
| 5 | 105 | 108 | -2.78% | 7.7160 |
Sum of
If the annualized return were 10% and the risk-free rate 2%, the UPI would be: .
Advantages & Limitations
Advantages
- Downside focus: Only measures declines from peaks, aligning with investor intuition about risk.
- Duration sensitive: Penalizes prolonged drawdowns more heavily than brief dips of the same magnitude.
- Easy to compute: Simple formula with no distributional assumptions or optimization required.
- Behavioral relevance:Captures the "pain" of sustained losses that drives real investor behavior and fund redemptions.
Limitations
- No probabilistic interpretation: Unlike VaR or CVaR, the Ulcer Index does not provide a probability statement about future risk.
- Sample dependent: Increases with the length of the observation period, making cross-period comparisons difficult.
- Not a coherent risk measure: Does not satisfy the formal axioms of coherent risk measures.
- Limited theoretical foundation: Primarily an empirical and heuristic measure without deep mathematical justification.
References
- Martin, P., & McCann, B. (1987). The Investor's Guide to Fidelity Funds.John Wiley & Sons.
- Bacon, C. R. (2013). Practical Risk-Adjusted Performance Measurement.John Wiley & Sons.
- Martin, P. (1987). "Ulcer Index: An Alternative Approach to the Measurement of Investment Risk and Risk-Adjusted Performance." Unpublished manuscript.
- Rollinger, T., & Hoffman, S. (2013). "Sortino: A Sharper Ratio." Red Rock Capital.