UMBC High Performance Computing Facility
Please note that this page is under construction. We are documenting the
240-node cluster maya that will be available after Summer 2014.
Currently, the 84-node cluster tara still operates independently,
until it becomes part of maya at the end of Summer 2014.
Please see the 2013 Resources Pages under the Resources tab for tara information.
Estimation of VARMA models
Anindya Roy, Mathematics and Statistics
Peter Linton, Mathematics and Statistics
Tucker McElroy, US Census Bureau
Maximum likelihood estimation of the VARMA model under causality constraints.
The standard likelihood evaluation via innovations algorithm is fast, but for
not fast enough for iterative calulations. Thus, for estimation it is
imperative that likelihood evaluations are done fast enough. The objective
will be to figure out a way to compute the likelihood within any iterative
scheme that would permit estimation in a reasonable time frame