As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
We propose a novel approach combining vector autoregressive models and data assimilation to conduct econometric inference for high dimensional problems in cryptocurrency markets. We label this new model TVP-VAR-DA. As the resulting algorithm is computationally very expensive, it mandates the introduction of a problem decomposition and its implementation in a parallel computing environment. We study its scalability and prediction accuracy under various specifications.
loading subjects...
