R: Monetary Value of Random Portfolios

valuation.randportBurSt {PortfolioProbe}

R Documentation

Monetary Value of Random Portfolios

Description

Returns one of various concepts of value for the portfolios.

Usage

## S3 method for class 'randportBurSt'
valuation(x, prices, weight = FALSE, 
	collapse = is.array(prices), type = "gross", cash = NULL, 
	all.assets = FALSE, returns = NULL, cumulative = FALSE)

Arguments

`x`	required. An object of class `"randportBurSt"` (generally the result of a call to `random.portfolio`).
`prices`	required. A named vector of asset prices that includes at least the assets in the object being evaluated; or a matrix with column names for the assets (rows are for times); or a three-dimensional array with rows corresponding to times, columns to assets, and slices to different scenarios.
`weight`	logical value; if `TRUE`, then the result will contain weights rather than monetary value.
`collapse`	logical value, or an object representing categories. If `TRUE`, then the result will be a vector or matrix containing results for the portfolios rather than individual assets. Categories can be represented by a named character vector or factor, or a list of these (for combinations of categories). The list can be in the form of a data frame where the row names are the asset identifiers.
`type`	this is only used when `collapse` is not `FALSE`. A character string that (partially) matches one of: `"gross"`, `"net"`, `"long"`, `"short"`, `"nav"`, `"cash"`. When this has value `"nav"`, then the result is the net value plus cash. When this has value `"short"`, the result is non-negative numbers. The `"cash"` type lets you see the value of cash that is computed.
`cash`	this is only used when `type="nav"` and `collapse` is not `FALSE` (which includes return calculations). A single number, a numeric vector with as many elements as there are rows in `prices`, or `NULL`. If this is `NULL`, then the amount of cash is such that the initial value is equal to the gross value minus the net value at the first time point for each portfolio.
`all.assets`	a logical value. If `TRUE`, then the `weight` and `individual` components will have the same assets (in the same order) as `prices`.
`returns`	either `NULL` or a single character string that is (an abbreviation of) `"log"`, `"simple"`, `"geometric"`, or `"arithmetic"`. This fails if `prices` is not a matrix or array. This is two types of returns with two names each.
`cumulative`	a logical value. If `TRUE`, then the returns are cumulative – that is, relative to the prices at the first time point. If `FALSE`, then the return at each time point is relative to the immediately previous time point. This is ignored if returns are not produced.

Value

a list, vector, matrix or three-dimensional array containing valuation information, and with some attributes.

If collapse is FALSE, then the result is a list of the same length as the input x where each component holds the asset valuations or weights.

If collapse is TRUE, then the result is an object with the same number of dimensions as prices, but the dimension in prices that corresponds to assets corresponds to random portfolios in the result.

If collapse represents categories, then the result has one more dimension than prices (and currently prices is not allowed to be three-dimensional in this case). The order of dimensions is: time (if any), categories, random portfolios.

The (extra) attributes are:

`timestamp`	character string giving the date and time of the evaluation.
`call`	the call that created the object.

Details

This is a method for the generic function valuation for class randportBurSt.

If the portfolios are long-short and the valuations are going to be used to calculate returns, then you should set type="nav" and set cash properly for the situation that you have.

If out.trade is TRUE in the call for x, then the results are for the trades and not for the portfolios.

Revision

This help was last revised 2013 April 17.

Examples

randport1 <- random.portfolio(100, prices, varian, long.only=TRUE,
	bench.constr=c(spx=.04^2/252), lin.constraints=cntrysect.conmat,
	lin.bounds=cntrysect.bounds, gross.value=1e6)

randval1 <- valuation(randport1, prices)
randval1n <- valuation(randport1, new.prices)

randret <- valuation(randport1, price.matrix, returns="simple")

# weights by country-sector combination
randwgt.cntrysect <- valuation(randport1, price.matrix, 
	collapse=list(country, sector), weight=TRUE)

[Package PortfolioProbe version 1.06 Index]