The presently observed negative interest rates for the recently emitted negative interest bonds by France, etc seem to increase in magnitude with the term. This might suggest that their modelling is not different from the one of "standard" positive rates, if one is not taking care of the sign, modelling just the magnitude.
Should the modelling of the observed negative interest rates be the same as for the observed positive ones?
(From the teoretical point of viw one can do it via imaginary rates, that become negative the moment they are observed, with a properly defined metrics and operators, but this is not actually within the topic of this question and does not lack modelling solutions).
