Direct answer: what data is needed to assess Hma
To assess Hma (Hull Moving Average) in a way you can independently verify, you need four groups of data: (1) the exact inputs to the calculation, (2) the provenance and consistency of the underlying price series, (3) the timeliness and alignment of that series with your chosen timeframe, and (4) quality checks that confirm the output matches the specified formula and parameters.
Mechanism or definition: what Hma needs to compute
Hma is defined by a specific moving-average procedure that transforms a raw price series into a smoothed line. Assessing it starts with the non-negotiable calculation inputs:
- Price data used as input (for example, the “close” price from candles). You must state which field is used, because changing the price source changes the result.
- Sampling timeframe (for example, 1-minute, 1-hour, or daily candles). The moving average operates on the sequence of data points, so timeframe is part of the assessment.
- Parameter values (commonly a length/period and any internal smoothing steps implied by the chosen Hma definition). Even when two sources both say “Hma,” parameter interpretation can differ.
- Data continuity rules (how the series treats missing bars, market open/close gaps, holidays, and weekend gaps). Different vendors handle these situations differently.
Stable mechanics vs variable conditions: the mechanics of “apply the formula to a time series with given parameters” is stable, but the market series itself, the provider’s construction of candles, and the way missing data is handled are variable.
Evidence or example: how to verify Hma outputs with the same data
A practical evidence-based way to assess Hma is to reproduce it from scratch using the same stated inputs.
Assumptions (must be explicit):
- You are using a fixed historical dataset (not real-time) so comparisons are deterministic.
- You will use one chosen definition of Hma and one chosen parameter set.
Verification checks:
- Provenance check: confirm where the price series comes from (data provider or platform), and ensure you use the same symbol and contract definition.
- Timeliness check (historical context): confirm the candle boundaries match the timeframe you claim (e.g., that each 1-hour bar covers the intended hour in the same time zone).
- Reproducibility check: compute Hma and compare the resulting values at several timestamps. If values differ, at least one of the inputs (price field, timeframe, parameters, or data handling rules) is likely different.
Independent verification also means you should be able to explain which inputs caused the differences when they appear.
Limitations and risks: material failure modes
Even if Hma is computed correctly, assessment can fail for several reasons:
- Stale or mismatched data: using an updated dataset, a different candle construction, or a different time zone can change the computed line.
- Inconsistent series: switching between close, typical price, or another derived input changes the meaning of the output.
- Parameter misunderstanding: “length” can be interpreted differently across implementations, leading to apparent agreement that is actually not equivalent.
- Lookback bias: relationships observed in one historical window do not establish future behavior; historical patterns can be coincidence.
- Cost and execution effects (if you later connect Hma to decisions): spreads, commissions, and fill timing can dominate what a line visually suggests.
A key limitation in this assessment context is that Hma is a derived indicator, not a guarantee of outcomes. Any conclusion about usefulness must acknowledge uncertainty and the dependence on the input data quality.
Verification or next question: what to document before trusting conclusions
Before you draw any conclusion from Hma, document the following in a checklist:
- Exact Hma definition used (and parameter values).
- Price field (e.g., close) and the symbol/contract definition.
- Timeframe and time zone handling for candle timestamps.
- Data source and any rules for missing or adjusted bars.
- A reproducibility test: confirm the line matches on several historical timestamps.
If two implementations disagree, the next question is not “which one is true” but “which input differs.” Typically, differences come from price source, timeframe alignment, parameter interpretation, or data continuity handling.