periods of high volatility cluster together, as do periods of low volatility.
About
A stylized fact about financial markets where volatile periods tend to be followed by other volatile periods.
This phenomenon is commonly observed in financial time series data.
Statistically, volatility clustering implies time-varying conditional variance: big volatility (variance) today may lead to big volatility tomorrow. This is why standard ARIMA models fail for financial returns—they assume constant variance.
The ARCH/GARCH family of models was specifically designed to capture this volatility clustering behavior by modeling the conditional variance as a function of past squared returns.
Interpretation
Intuitively, the market becomes volatile whenever big news arrives, and it may take several periods for the market to fully digest the news. This creates the observed clustering pattern where a shock today leads to higher uncertainty tomorrow.