Investigating Volatility Around Options Expiration
An expiration-week pattern is a hypothesis to examine, not a property this tool assumes. The calculation uses observed historical closing prices to compare realized volatility around a recurring calendar marker. It does not manufacture past option prices or infer a market-making position from the calendar.
How to Use the Comparison
- Enter a stock or ETF symbol and choose a one-, two-, or three-year lookback.
- Select a 10-, 20-, or 30-return rolling RV window. Click Compare RV. Changing any input clears the prior result.
- Inspect the historical chart, then the paired-window averages and observation counts. The third-Friday week is Monday through Friday; its comparison window is the immediately preceding Monday through Friday.
- Check missing observations and the returned price range. The current UTC date is excluded to avoid using an incomplete daily close. Results are not live market signals.
What the Realized Volatility Formula Means
For each consecutive supplied closing price, the return is ln(close today) minus ln(close previously). For N returns, subtract their mean, square the deviations, sum them and divide by N − 1. The square root is multiplied by √252 and 100 to display annualized RV in percent. The 252 factor is a trading-day convention, not a forecast of actual future volatility.
A weekly value is the arithmetic mean of the available rolling RV readings in that calendar window. It is not volatility calculated only from the returns inside that week. The summary gives each matched week pair equal weight. A target average of 25% and prior average of 23% differ by 2 percentage points, not 2% relative to the prior value.
Only finite, positive closes with valid dates are accepted. Conflicting duplicates fail visibly. Gaps longer than seven calendar days reset the rolling calculation. Shorter missing sessions may still affect the estimate; the feed is not audited against a full exchange-session calendar. Review provider adjustments and corporate actions before interpreting extreme changes.
Calendar Weeks Are Not Verified Expiration Events
The third-Friday marker is a transparent proxy. Actual contracts can expire on other weekdays, and holidays can move monthly dates. For example, the OCC 2026 schedule lists June 18, not the third Friday June 19, as that month’s expiration. This tool keeps the third-Friday calendar group and does not claim to identify every actual expiry.
Only completed calendar windows in the selected lookback are compared. Missing windows remain unavailable, and the prior-week group is not every non-expiration week in the year. Use the expiration calendar and exchange specifications to research a specific contract.
Why Historical IV and Significance Are Not Shown
Historical implied volatility requires option observations and a consistent contract-selection method at each historical date. A present-day snapshot cannot supply that history. Realized volatility multiplied by a guessed premium is not an observed IV series.
Rolling RV estimates overlap and therefore share inputs. The comparison weeks are paired and financial time series can exhibit dependence. A simple independent-samples test would not by itself establish a reliable expiration effect. This page reports descriptive differences without p-values, causal conclusions or a buy/sell recommendation.
Sources and Further Checks
- FMP historical-price documentation for the underlying close data.
- OCC 2026 holiday and expiration schedule for actual monthly dates.
- NIST autocorrelation guidance for dependence in time series.
Compare results across windows and market regimes before forming a hypothesis. A trading evaluation would also need option prices, costs, risk limits and out-of-sample testing. Historical differences alone do not establish profitability or suitability.