With a VAR your expected returns are changing at each time period. You could do 
a historicalsimulation where the the MVP portfolio is computed at each time 
step based on the expected1-period returns and the covariances. There is more 
evidence that covariances are predictablethan expected returns, so ideally you 
will have a model for time-varying covariances, maybe something as simple as an 
EWMA.
Transaction costs may negate the benefits of updating the portfolio often. One 
simple way toreduce transaction costs is to trade X% (say 10%) of the way 
toward the target each day.
In general, this is a dynamic programming problem. With an estimated VAR, you 
can predictnot just 1-period but N-period returns through iteration, and people 
have thought about how to optimize allocations when you have return forecasts 
over various time horizons. (Are thereR packages for this?)
Regards,Vivek Rao, CFABoston, MA
      From: "Fianu, Emmanuel Senyo" <emmanuel.se...@gmail.com>
 To: r-sig-finance@r-project.org 
 Sent: Wednesday, July 11, 2018 7:16 AM
 Subject: [R-SIG-Finance] Mean Variance Portfolio Optimization based on a DGP
   
Dear All,

I am trying to employ the MVP method to determine optimal weights for a
data, which is fine for me to do.
However, I intend to use a different Data generating process (DGP) such as
Vector Autoregressive  Process (VAR), and then compute the optimal weights.
Theoretically, it looks okay but empirically, have someone carried out this
before? If yes: how did you go about it?

 I would be grateful for your constructive and helpful comments.

Many thanks,
Emmanuel Fianu

    [[alternative HTML version deleted]]

_______________________________________________
R-SIG-Finance@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should 
go.


   
        [[alternative HTML version deleted]]

_______________________________________________
R-SIG-Finance@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should 
go.

Reply via email to