Aggregating to a weekly scale smooths out short-lived fluctuations while retaining the
most significant geomagnetic disturbances, as well as reducing short-term serial dependencies.
The extreme nature of the data motivates modelling via the Generalised Extreme Value (GEV)
distribution, with density
$$f(x) = \frac{1}{\sigma}
\left[1+\xi\left(\frac{x-\mu}{\sigma}\right)\right]^{-\frac{1}{\xi}-1}
\exp\left\{-\left[1+\xi\left(\frac{x-\mu}{\sigma}\right)\right]^{-\frac{1}{\xi}}\right\},$$
where \(\mu\), \(\sigma\), and \(\xi\) are the location, scale, and shape parameters, respectively.