| mergexy {costat} | R Documentation |
Concatenate a set of solution results into one set
Description
Merges several sets of optimization results from
multiple calls to findstysols into
a single object for further analysis
Usage
mergexy(...)
Arguments
... |
An unspecified number of arguments of class |
Details
The return object from an invocation of the
findstysols is a list containing a number
of interesting components containing information about
the starting parameters, the (hopefully optimal) ending
parameters, convergence status, minimum variance achieved
and p-value associated with the final test of stationarity
after an optimization.
It is possible to ask findstysols to execute
multiple optimization runs in the same function, by choice of
the Nsims parameter. However, for truly large runs,
it can be convenient to run multiple copies of
findstysols, for example on multiple processors
simultaneously (a coarse grained parallelism).
In particular, for large time series, it can be useful to run
findstysols for one optimization run
(as running more than one for a very large series can cause the
software to fail as R can run out of memory. Actually, for very
very large series even one optmization run can fail for memory
reasons).
In this way multiple optimization runs can be executed with each
one producing its own set of results. This function
(mergexy) takes a list of object names of all of the results,
and merges the results into one object as if a single call
to findstysols had been executed. Such a single
set of results can then be passed on to further analysis
routines, such as COEFbothscale or
LCTSres.
Value
A set of optimization solutions in the same format as
those returned by findstysols
Author(s)
Guy Nason
References
Cardinali, A. and Nason, Guy P. (2013) Costationarity of Locally Stationary Time Series Using costat. Journal of Statistical Software, 55, Issue 1.
Cardinali, A. and Nason, G.P. (2010) Costationarity of locally stationary time series. J. Time Series Econometrics, 2, Issue 2, Article 1.
See Also
findstysols, LCTSres,
COEFbothscale
Examples
#
# Generate two dummy time series
#
x1 <- rnorm(32)
y1 <- rnorm(32)
#
# Run two optimizations
#
## Not run: solnset1 <- findstysols(Nsims=1, tsx=x1, tsy=y1)
## Not run: solnset2 <- findstysols(Nsims=1, tsx=x1, tsy=y1)
#
# Merge them
#
## Not run: solnset <- mergexy(solnset1, solnset2)