Regime-shifts seem to affect price deviations but not stationarity?

General
1

Course Name: Financial Time Series Analysis for Trading, Section No: 17, Unit No: 9, Unit type: Notebook

I am a bit confused. While the mean of the time-series indeed seems to be constant for
data['Difference'] = data['Close'].diff()
the deviations away from it might not.

I seem to identify two regimes:
Between 2012 to 2016 the deviations are between ± 2 and from 2016 to 2021 they are ±0.5.

While the deviations away from the mean may be considered symmetric within these two regimes, at least visually, I would not say they are constant.

Could you please further elaborate why the deviations are evaluated to be constant in time?

2

Hi Benjamin, we are looking into your questions with regards to the material. Will get back to you shortly.

3

Hi Benjamin,
Thanks for your observation.

You are right that the variability appears to differ across periods, with larger deviations in the earlier years and smaller ones later. The intention of the discussion was to highlight the symmetry of deviations around the mean, rather than strict constancy of variance.