Distribution fitting in Python: Generalised error distribution
It is very common for real-world asset return distributions to have tails much fatter than predicted by the normal distribution yet much thinner than prescribed by pathological distributions such as Cauchy. To address that, many flexible generalised distribution families have been developed. One of the most famous and most widely applied is the generalised error (or generalised normal distribution). Today we are investigating the mathematics and the conceptual background behind it and building a flexible algorithm in Python to optimise it using maximum likelihood and evaluate its goodness-of-fit for real-world financial markets.
Don't forget to subscribe to NEDL and give this video a thumbs up for more videos in Python!
Please consider supporting NEDL on Patreon: / nedleducation