R: Cache a set of variance matrices

pp.cacheVar {pprobeSup}

R Documentation

Cache a set of variance matrices

Description

Creates and stores a series of variance matrices estimated from the given return matrix.

Usage

pp.cacheVar(cache, times, retmat, lookback = 250, offset = 0, 
	prefix = "var", FUN = var.shrink.eqcor, verbose = TRUE, 
	...)

Arguments

`cache`	a character string giving the name of the directory in which to store the files.
`times`	a vector of elements of the row names of `retmat` at which estimates of the variance are desired. The `lookback` and `offset` arguments limit the earliest time that is possible.
`retmat`	a matrix of returns for the assets with rows that include the range of times for which variances are desired.
`lookback`	a single integer giving the number of rows of `retmat` to use in each estimation.
`offset`	a single integer giving the number of rows to offset in the estimation. The default of 0 means that the time on the object containing a variance uses the returns with that time. A value of 1 would mean that the latest returns used for an object are 1 time before the time on object name.
`prefix`	a character string giving the prefix to use in the object names. The rest of the name is the corresponding element of `times`.
`FUN`	the function to estimate the variance.
`verbose`	logical value; if `TRUE`, then the progress is reported as each evaluation of `FUN` is completed.
`...`	additional arguments may be passed to `FUN`.

Details

While this was written with the intention of it being used to store variance matrices, there is no reason why FUN needs to return a variance matrix. The reason to use cacheing is if either the output is large and/or the computation takes a reasonably long time and the object will be used multiple times.

Value

NULL is returned invisibly.

Side Effects

R objects are saved into the cache directory with suffix rda.

Testing status

Not formally tested.

Revision

This help file was last revised 2013 May 02.

Examples

require(BurStFin)
require(pprobeData)
yearend <- rownames(xassetLogReturns)[cumsum(table(
	substr(rownames(xassetLogReturns), 1, 4)))]
pp.cacheVar("VarianceStash", yearend, xassetLogReturns)

[Package pprobeSup version 1.00 Index]