The Allan variance, also known as two-sample variance, is a measurement of accuracy in clocks. It is defined as one half of the time average over the sum of the squares of the differences between successive readings of the frequency deviation sampled over the sampling period. For most real-world systems, the Allen variance depends on the time period used between samples: therefore it is a function of the sample period, as well as the distribution being measured. A low Allan variance is a characteristic of a clock with good stability over the measured period.
The Allan variance is conventionally expressed by σy2(τ). The samples are taken with no dead-time between them.
Source: from Federal Standard 1037C