Logarithmic Garman–Klass Volatility Estimation from OHLC Prices

Summary

The document presents a logarithmic Garman–Klass estimator that uses open, high, low, and close prices to estimate realized or historical volatility. It forms log returns for close-to-close, close-to-open, open-to-previous-close, and high-to-low moves. The variance input combines the overnight open gap and the intraday range, with a negative adjustment based on the close-to-open term. The result is smoothed in log space using a weight derived from the lookback length, then scaled by a compounding factor and expressed as a percentage.

The estimate is initialized from close-to-close historical variance. The author claims it responds quickly while avoiding some overshooting and lag associated with other approaches, but provides no comparative dataset, parameter study, or performance results to substantiate that claim. The formula requires careful implementation of the lookback, compounding period, and price inputs; the post does not discuss data quality, market-specific suitability, or forecast accuracy. It is an indicator estimate, not by itself a trading signal.

Key ideas

• The estimator uses logarithmic price changes derived from OHLC data.
• Its variance formula combines the overnight price gap with the intraday high-low range.
• A lookback-dependent weight smooths the variance estimate in logarithmic space.
• The final volatility value is scaled using a compounding factor and reported as a percentage.
• The document gives no empirical comparison or evidence for the claimed response characteristics.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.