[R-SIG-Finance] Partitioning approach towards portfolio construction - ensemble models/weights based on parameter sets

Enrico Schumann e@ @end|ng |rom enr|co@chum@nn@net
Wed Jul 24 08:03:36 CEST 2019


>>>>> "Sam" == Sam H <sam.hhh1 using gmail.com> writes:

    Sam> Hi,
    Sam> Is there some (example) code available somewhere (can be highly
    Sam> experimental) that would enable conducting this kind of analysis (portfolio
    Sam> construction) (possibly wrapping PortfolioAnalytics):
    Sam>     - https://blog.thinknewfound.com/2019/07/ensemble-multi-asset-momentum/
    Sam>     -
    Sam> https://docs.wixstatic.com/ugd/7c4c63_b3f66bbea0f648e19e535b1da004aeba.pdf
    Sam>     -
    Sam> https://docs.wixstatic.com/ugd/7c4c63_735bc38a987340cc8db85691a41dbfe4.pdf

    Sam> So to be able to create average/ensemble weights based on a set of
    Sam> parameters (like rebalance date, look back periods for momentum and
    Sam> whatever the parameters are). Something like quantstrat has
    Sam> with apply.paramset, add.distribution, add.distribution.constraint, ...

    Sam> Original message was not delivered due to attachments, I guess.
    Sam> -- 
    Sam> Best regards,
    Sam> Sam

Perhaps the examples in https://ssrn.com/abstract=3374195 are of
interest (though they do not use PortfolioAnalytics).

kind regards
     Enrico

-- 
Enrico Schumann
Lucerne, Switzerland
http://enricoschumann.net



More information about the R-SIG-Finance mailing list