Glossary
Autocorrelation
The correlation of a series with its own past values at a chosen lag.
Autocorrelation at lag k is the correlation between a series and the same series k periods earlier. Positive values mean moves tend to continue; negative values mean they tend to reverse. For n independent observations, a sample autocorrelation has a standard error of about 1 / sqrt(n), so small values are often indistinguishable from noise.
Daily returns usually show little autocorrelation, while squared or absolute returns show a lot (volatility clustering).
Example: with 1,000 returns, the 95% noise band is plus or minus 1.96 / sqrt(1000) = 6.20%. A measured lag-1 autocorrelation of 4% lies inside it.