Direct answer
Ulcer Index is sensitive to the timeframe because it summarizes drawdowns that occur within the observation window. If you change the start/end dates (or the sampling frequency) you change which highs, drops, and recovery depths are included, so the Ulcer Index can move up or down even when nothing else is “re-calculated” in the real world.
Mechanism or definition
Ulcer Index is a drawdown-based measure. In practice it focuses on how far an observed value falls below its earlier peak, and it aggregates those drawdown depths over the selected period.
Two timeframe-related inputs matter:
- Observation window (holding period / lookback range). A longer window has more opportunities to experience drawdowns and recoveries. That alone can increase the average severity of drawdown depths included in the calculation.
- How you sample the series within the window. If you compute it using more frequent observations (for example, many small steps instead of a few), you may capture more interim highs and lows. That changes the sequence of peak-to-trough distances.
What “timeframe affect” means for interpretation
When you read a higher Ulcer Index for a given timeframe, it means: within that window, the series spent more of its time in larger drawdowns (and/or reached deeper troughs) than a comparison timeframe. When you read a lower Ulcer Index, it means the opposite for the same type of data and method—but only for the specific window you chose.
Evidence or example
Assume you compute Ulcer Index from a price or performance series using a defined method: you track the running maximum within the window and compute drawdown depths relative to that maximum; then you aggregate those depths into a single number.
Scenario (illustrative, with explicit assumptions):
- Same underlying pattern, different windows.
- Window A covers a period that includes one sharp drop and recovery.
- Window B covers only the quiet part before the drop.
If Window B never includes that sharp drop, the drawdown depths included in the aggregation are smaller or near zero most of the time. As a result, the Ulcer Index for Window B will likely be lower than for Window A.
Scenario (sampling within a window):
- You keep the same start/end dates but use a coarser sampling approach for Window C.
- If the coarser series skips the exact trough moments, the measured drawdown depths can be less extreme than if you had sampled more densely.
These examples show the core sensitivity: Ulcer Index reflects what happened in the data slice you feed into the computation, not an intrinsic property of the market that stays constant across all possible timeframes.
Limitations and risks
- Comparability depends on matching methods. Different sampling frequencies, different calculation definitions, or different window choices can produce different values that are not directly comparable.
- Window selection can change conclusions. A timeframe can unintentionally emphasize rare events (large spikes in drawdowns) or exclude them, shifting Ulcer Index even if typical behavior is similar.
- Failure mode: missing extremes. If the sampling interval is too coarse, measured drawdowns can understate the true peak-to-trough depth, making Ulcer Index look artificially low.
- No predictive guarantee. Ulcer Index summarizes historical drawdown behavior in the selected period. It does not, by itself, guarantee future drawdown behavior or timing.
Verification or next question
To independently verify the “timeframe sensitivity” idea, recompute Ulcer Index on the same dataset while varying only the timeframe (and keeping the calculation method constant). If the value changes materially, that supports the interpretation that the observation window affects which drawdowns are included.
A helpful next question is: What exact sampling and calculation choices were used for the timeframe you are comparing? Even with the same dates, changing sampling frequency or aggregation rules can alter drawdown depth estimates and therefore the Ulcer Index result.