Q&A

The following are general questions which provide you overview of the usage and limitation of the library.

Is the risk model annualised?

The granularity of the risk model depends on the instrument returns fitting the risk model. If you input daily returns into the method fit, the covariance matrix returned by the same risk model contains the pairwise daily covariance entries, while the correlation is unaffected. In short, you need to manually annualise your risk model regarding the granularity of fitting data.

What are the expected input types?

Currently only Pandas (DataFrame / Series) and NumPy (array) are accepted in modeling input, while we are extending to accept Polars (DataFrame / Series) in the future.