New and Improved Volatility Oracle Deployed

tl:dr; The audit over the new Volatility Oracle was successfully completed and the contract has been deployed to mainnet.

The new Volatility Oracle is already being used to facilitate trading of options on the Lien platform. This is an extremely important component and a major leap in improving option pricing on decentralized protocols. The audit report can be reviewed here.

What is the Volatility Oracle and how was it improved

The original Volatility Oracle calculated volatility based on the latest 24 data points from the Chainlink price feed.

Technically, implied volatility is calculated based on option prices in the market and not from historical prices of the underlying. Unfortunately, there are no Oracles, or similar services that can reliably provide such implied volatility data. Therefore, as a temporary measure, we are using historical volatility based on the Chainlink Price Oracle data when pricing options on Lien.

The Chainlink Price Oracle generates a new data point when the price of underlying assets(e.g. ETH, BNB) moves approximately 0.5% or more, or at least 1 hour. When the market is volatile, historical volatility is calculated based on the 24 data points during the period when the price is moving the most. This is primarily the reason why we were witnessing multiple volatility spikes exceeding 300% on the platform.

On the other hand, when volatility subsides and the market stabilizes, users would witness an abrupt drop in volatility as the price data during the high volatility period is kicked out of the 24 data point slots and replaced with subsequent price data.

The duration of the options sold on Lien extend over a period of a few weeks. A few weeks is a long time, long enough where there are days when the market is extremely volatile and there are days when the market is calm and steady. Naturally, the volatility used in pricing the options should reflect the expected price behaviour over that time period.

This was a challenge that needed to be addressed for a better user experience.

With the new and upgraded Volatility Oracle, we have implemented the concept of Exponentially Weighted Moving Average or EWMA.

EWMA gives more weight to recent data compared to older data.

The result is a smoother transition of volatility during periods of significant price swings and relative stability.

We also made smaller tweaks to improve the efficiency of certain calculations.

For example, from now on, the Volatility Oracle stores Chainlink price data every day on UTC0:00 and will use that as the regular interval daily price data.

This improves the consistency of intervals between data points as the Chainlink price data is not updated in consistent intervals, as noted above.

What’s Next

There’s another contract that has be audited and this one is smoking…