ML//Time Series//point process

A model for the random timing of discrete events on a line (or in space), rather than a value sampled on a fixed clock.

A model for the random timing of discrete events on a line (or in space), rather than a value sampled on a fixed clock.

The baseline is the Poisson process: events arrive independently at a constant rate.

A Hawkes process adds self-excitation; other variants add inhibition or external drivers.

You model the intensity (the instantaneous rate), not the events directly.