percentrank_tdigest()

Learn how to use the percentrank_tdigest() function to calculate the approximate rank of the value in a set.

Calculates the approximate rank of the value in a set, where rank is expressed as a percentage of the set’s size. This function can be viewed as the inverse of the percentile.

Syntax

percentrank_tdigest(digest, value)

Parameters

NameTypeRequiredDescription
digeststring✔️An expression that was generated by tdigest() or tdigest_merge().
valuescalar✔️An expression representing a value to be used for percentage ranking calculation.

Returns

The percentage rank of value in a dataset.

Examples

Getting the percentrank_tdigest() of the damage property that valued 4490$ is ~85%:

StormEvents
| summarize tdigestRes = tdigest(DamageProperty)
| project percentrank_tdigest(tdigestRes, 4490)

Output

Column1
85.0015237192293

Using percentile 85 over the damage property should give 4490$:

StormEvents
| summarize tdigestRes = tdigest(DamageProperty)
| project percentile_tdigest(tdigestRes, 85, typeof(long))

Output

percentile_tdigest_tdigestRes
4490