Forex Volatility: Why the Observation Window Matters

Volatility research · Published 2026-10-03

In 2024’s frozen reference sample, twenty-return annualised EUR/USD volatility ranged from 3.08% to 10.99%, while USD/JPY ranged from 4.10% to 23.98%. These estimates describe different windows of observed returns; they are not forecasts, price ranges or annual trading outcomes.

Original analysis by InsomniCapital. The observation period or hypothetical assumptions below determine what this study can establish; its publication date does not make the inputs live.

The same pair has more than one historical volatility estimate.

EUR/USD’s lowest twenty-return estimate ends on 13 March 2024 and its highest on 26 November. USD/JPY’s extremes end on 24 April and 7 August. Their maxima occur in different parts of the year. Reporting only a full-year figure would conceal that timing, while selecting only the largest short-window estimate would exaggerate what the rest of the sample looked like. Neither approach answers the question of how the estimate changes as observations arrive.

Changing the lookback changes the question.

The sixty-return calculation uses more observations at each endpoint and produces a different range of estimates. Its first eligible endpoint is later in the year because sixty-one prices are required. Consequently, comparing the overall minimum or maximum of the two lookbacks also compares different eligible endpoint sets. To isolate a same-date comparison, select the twenty-return and sixty-return rows with exactly the same last_date value in the CSV.

We chose twenty and sixty consecutive reference intervals as transparent teaching examples. They are not optimised settings, fixed calendar months or a recommendation to resize positions at either frequency. Shorter windows can react more sharply to a small number of large returns; longer windows retain earlier movements for longer. The data does not identify a universally correct lookback.

Estimate dispersion from consecutive log returns.

For each pair and eligible endpoint, select the preceding twenty-one reference dates, including that endpoint. Form twenty returns as ln(current reference rate ÷ previous reference rate). Subtract their mean, square each deviation, sum the squares, divide by nineteen and take the square root. Multiply the resulting sample standard deviation by √252 to obtain the annualised convention displayed here. For sixty returns, use sixty-one prices and divide the squared-deviation sum by fifty-nine.

EUR/USD, EUR/GBP and EUR/JPY use their source columns directly. GBP/USD equals USD per EUR divided by GBP per EUR; USD/JPY equals JPY per EUR divided by USD per EUR. We derive each cross before calculating its return series. There are 236 twenty-return and 196 sixty-return windows per pair, with no rounding before the final display. JSON volatility is a decimal ratio; CSV columns explicitly multiply it by one hundred and use a percent suffix.

Volatility is neither direction nor a stop distance.

A 10% annualised estimate does not mean a currency pair moved 10% in this window, that its next-year return will be 10%, or that its price remained within a 10% band. Standard deviation describes dispersion around the sample mean. The square-root-of-time multiplier changes the unit convention; it does not create an observed annual outcome or a forecast. Both positive and negative deviations contribute to the result.

These once-per-publication reference observations cannot measure intraday high-low ranges, the path to an order, or whether a stop would have triggered between references. They therefore cannot replace ATR computed from actual high-low-close bars, broker contract specifications or an execution-aware test. Comparing a JPY pair with a non-JPY pair in percentage-return terms avoids raw pip-unit differences, but does not make their transaction costs, liquidity or event risks equivalent.

Use the daily reference dashboard to explore a separately dated recent sample, and the rolling correlation study to distinguish co-movement from each pair’s own dispersion. Higher volatility and stronger correlation are different properties and can change independently.

Reproduce the calculation from the source observations.

Download the frozen source CSV and this study’s result CSV and JSON. The input has 256 dates and three quotes per date: USD, GBP and JPY units per one EUR. The SHA-256 checksum is febf2b45b33628f8e8aceae6d28ced5833e78e4b393b13f14d5d316b8ee61bd8. The JSON records the publisher, original source query, retrieval timestamp, period, formulas and units. The result CSV preserves full floating-point output rather than the rounded numbers displayed in tables.

Open the source CSV in the historical forex lab. Select the two pairs and the exact first and last dates of a result row. The lab uses the same consecutive-observation convention. A twenty-return row must contain twenty-one reference dates; selecting twenty dates answers a different question. Compare the displayed rounded values, then inspect the downloads when greater precision is required. A later source revision must be reviewed and documented before replacing the frozen inputs.

What this sample cannot establish.

We use 2024 because it is the site’s existing reviewed full-year dataset, not because it was chosen from many years to support a trading claim. One year, three source currencies and derived crosses cannot represent every market regime. Adjacent rolling windows share most of their returns, so a count of windows is not a count of independent experiments. We report descriptive estimates without confidence intervals, causal claims or a statistical test of a permanent relationship.

The ECB publishes reference rates for information and discourages using them for transactions. They are neither a broker’s closing price nor intraday OHLC bars. A calendar gap remains a single observation interval; we do not insert weekends, holidays or interpolated values. The square-root-of-252 convention does not correct for unequal elapsed time, volatility clustering or serial dependence. These figures exclude execution costs and do not tell you the outcome of any position.

A small historical estimate does not establish safety. A large estimate does not identify the direction of the next move. Decisions about a demo exercise still need an explicit invalidation rule, contract units, combined exposure and an allowance for costs. Read the position-sizing-with-costs lesson before converting a hypothetical stop distance into quantity.

Publication and corrections

• 2026-10-03 — Initial publication.

No corrections recorded for this version. Substantive corrections should describe the changed claim, its effect on the findings and any replacement download. Read our research and corrections method.

Downloads and primary sources

• All 432 windows across five pairs (CSV)
• Results, source provenance and methods (JSON)
• Original rolling volatility chart (SVG)

The downloads contain the inputs, units or methods required to understand the results. Check the source period and assumptions before reusing them.

• European Central Bank: reference-rate purpose and publication convention
• ECB Data Portal: official 2024 USD, GBP and JPY source query
• European Central Bank: reuse and attribution terms

Source checks and arithmetic checks do not establish a profitable trading strategy. No independent professional review, live trading result or forecast is claimed. Read the risk disclosure.