| update.randportBurSt {PortfolioProbe} | R Documentation |
Re-executes a call to random.portfolio, possibly with some arguments
changed.
update.randportBurSt(object, ..., evaluate = TRUE,
checkinput = TRUE)
object |
an object produced by random.portfolio
(or random.portfolio.utility).
|
... |
arguments to random.portfolio that should be
added or changed from those in the initial call.
|
evaluate |
logical value.
If TRUE (the default), then the command is executed.
If FALSE, then the (revised) call is returned.
This argument needs to be given by its full name since it comes after the three-dots in the argument list. |
checkinput |
logical value.
If TRUE (the default), then a check is made to see
if the prices, variance, expected return and the existing portfolio
are the same in the new call as the original call.
Except that any of these are excluded from the test if they
are arguments in the call to update.
This argument needs to be given by its full name since it comes after the three-dots in the argument list. |
If evaluate is TRUE, then the possibly revised
call is executed, producing a random portfolio object (of class
"randportBurSt").
Arguments that are changed need to use the exact same abbreviation (if any).
For instance, if in the original call net.value is abbreviated to
net.val, then you need to use net.val if you want to change
the net value allowed.
The default method of update works for the results of
trade.optimizer so it doesn't need a special method.
However, this means that there is no check for the key inputs changing.
If you want such a check, then you need to do it yourself as can be
seen in the help file for pprobe.checkinput.
This help was last revised 2009 December 29.
random.portfolio, trade.optimizer,
pprobe.checkinput.
randport1 <- random.portfolio(100, prices, variance=varian,
long.only=TRUE, bench.constr=c(spx=.04^2/252),
lin.constraints=cntrysect.conmat,
lin.bounds=cntrysect.bounds, gross.value=1e6)
randport2 <- update(randport1) # same again
randport3 <- update(randport1, gross.value=1.5e6)
opt1 <- trade.optimizer(prices, variance=varian,
long.only=TRUE, bench.constr=c(spx=.04^2/252),
lin.constraints=cntrysect.conmat,
lin.bounds=cntrysect.bounds, gross.value=1e6)
opt2 <- update(opt1, gross.value=1.5e6)
call1 <- update(opt1, number.rand=100, evaluate=FALSE)
call1[[1]] <- as.name("random.portfolio")
randport4 <- eval(call1) # random portfolios like opt1