Exchange Data Filtering

Exchange Data Filtering is the practice of identifying and excluding suspicious, low-quality, or erroneous trade data from index calculations. In the crypto space, some exchanges may report fake volume or experience technical glitches that produce outlier prices.

Filtering mechanisms use statistical tests to detect these anomalies and remove them before they affect the reference price. This ensures that the resulting index remains clean and representative of legitimate market activity.

Effective filtering is a key component of index methodology transparency, as it builds confidence among users that the data is not being skewed. It requires constant updates to the filtering criteria to adapt to new types of market manipulation.

This process is essential for maintaining the integrity of derivatives that rely on external price feeds.