One source of timing irregularity in middle aged pulsars is variability in the rate at which the pulsar slows down. A number of pulsars have been shown to oscillate between two different spin-down rates.
Pulsars slow down through the loss of rotational kinetic energy as their co-rotating magnetic fields imposing a braking torque. If the distribution of plasma in the magnetic field changes, so does the braking torque. This will lead to a change to the spin-down rate.
I’ve been working on simulations of pulsars with oscillating spin-down rates. Specifically I’ve been investigating the resolution possibilities of switching events in pulsar timing data. I’ll be shortly submitting this work for publication.
More recently I’ve been working on using Gaussian Process Regression to model spin-down and emission variations in Lovell Telescope data.