A discrete log-normal process to generate a sequential multifractal time series,
Résumé
We propose a discrete log-normal process to generate an intermittent time series and show that it possesses multifractal properties. With such a process, it is possible for the first time to sequentially generate a multifractal time series with an arbitrary length and a specified and fixed scale ratio. We explore its predictability properties and discuss a log-stable generalization. Since this corresponds to a special type of log-FARIMA truncated process, this model provides for the first time a link between FARIMA and multifractal families of models.