repRisk: Decompose portfolio risk into individual factor contributions...

In AvinashAcharya/Test_factorAnalytics: Development of factorAnalytics package as a part of GSoC-2016

Description

Compute the factor contributions to standard deviation (SD), Value-at-Risk (VaR), Expected Tail Loss or Expected Shortfall (ES) of the return of individual asset within a portfolio return of a portfolio based on Euler's theorem, given the fitted factor model.

Usage

repRisk(object, ...)

## S3 method for class 'tsfm'
repRisk(object, weights = NULL, risk = c("Sd", "VaR", "ES"),
  decomp = c("RM", "FMCR", "FCR", "FPCR"), digits = NULL, invert = FALSE,
  nrowPrint = 20, p = 0.95, type = c("np", "normal"),
  use = "pairwise.complete.obs", bystock = TRUE, isPrint = TRUE,
  isPlot = TRUE, ...)

## S3 method for class 'ffm'
repRisk(object, weights = NULL, risk = c("Sd", "VaR", "ES"),
  decomp = c("RM", "FMCR", "FCR", "FPCR"), digits = NULL, invert = FALSE,
  nrowPrint = 20, p = 0.95, type = c("np", "normal"), bystock = TRUE,
  isPrint = TRUE, isPlot = TRUE, ...)

Arguments

object	fit object of class tsfm, or ffm.
...	other optional arguments passed to quantile and optional arguments passed to cov
weights	a vector of weights of the assets in the portfolio, names of the vector should match with asset names. Default is NULL, in which case an equal weights will be used.
risk	one of 'Sd' (standard deviation), 'VaR' (Value-at-Risk) or 'ES' (Expected Tail Loss or Expected Shortfall for calculating risk decompositon. Default is 'Sd'
decomp	one of 'RM' (risk measure), 'FMCR' (factor marginal contribution to risk), 'FCR' 'factor contribution to risk' or 'FPCR' (factor percent contribution to risk). Default is 'RM'
digits	digits of number in the resulting table. Default is NULL, in which case digtis = 3 will be used for decomp = ('RM', 'FMCR', 'FCR'), digits = 1 will be used for decomp = 'FPCR'. Used only when isPrint = 'TRUE'
invert	a logical variable to choose if change VaR/ES to positive number, default is False
nrowPrint	a numerical value deciding number of assets/portfolio in result vector/table to print. Used only when isPrint = 'TRUE'
p	confidence level for calculation. Default is 0.95.
type	one of "np" (non-parametric) or "normal" for calculating VaR & Es. Default is "np".
use	an optional character string giving a method for computing factor covariances in the presence of missing values. This must be (an abbreviation of) one of the strings "everything", "all.obs", "complete.obs", "na.or.complete", or "pairwise.complete.obs". Default is "pairwise.complete.obs".
bystock	a logical value to choose slice/condition by stock(TRUE) or factors(FALSE. Default is TRUE. Used only when isPlot = 'TRUE'
isPrint	logical variable to print numeric output or not.
isPlot	logical variable to generate plot or not. isPlot = FALSE when decomp = 'RM'.

Value

A table containing

decomp = 'RM'	length-(N + 1) vector of factor model risk measure of portfolio return as well assets return.
decomp = 'FMCR'	(N + 1) * (K + 1) matrix of marginal contributions to risk of portfolio return as well assets return, with first row of values for the portfolio and the remaining rows for the assets in the portfolio, with (K + 1) columns containing values for the K risk factors and the residual respectively
decomp = 'FCR'	(N + 1) * (K + 2) matrix of component contributions to risk of portfolio return as well assets return, with first row of values for the portfolio and the remaining rows for the assets in the portfolio, with first column containing portfolio and asset risk values and remaining (K + 1) columns containing values for the K risk factors and the residual respectively
decomp = 'FPCR'	(N + 1) * (K + 1) matrix of percentage component contributions to risk of portfolio return as well assets return, with first row of values for the portfolio and the remaining rows for the assets in the portfolio, with (K + 1) columns containing values for the K risk factors and the residual respectively

Where, K is the number of factors, N is the number of assets.

Author(s)

Douglas Martin, Lingjie Yi

See Also

fitTsfm, fitFfm for the different factor model fitting functions.

portSdDecomp for factor model Sd decomposition. portVaRDecomp for factor model VaR decomposition. portEsDecomp for factor model ES decomposition.

Examples

# Time Series Factor Model
data(managers)
fit.macro <- factorAnalytics::fitTsfm(asset.names=colnames(managers[,(1:6)]),
                     factor.names=colnames(managers[,(7:9)]),
                     rf.name=colnames(managers[,10]), data=managers)
report <- repRisk(fit.macro, risk = "ES", decomp = 'FPCR', 
                  nrowPrint = 10)
report 

# plot
repRisk(fit.macro, risk = "ES", decomp = 'FPCR', isPrint = FALSE, 
        isPlot = TRUE)

# Fundamental Factor Model
data("stocks145scores6")
dat = stocks145scores6
dat$DATE = as.yearmon(dat$DATE)
dat = dat[dat$DATE >=as.yearmon("2008-01-01") & 
          dat$DATE <= as.yearmon("2012-12-31"),]

# Load long-only GMV weights for the return data
data("wtsStocks145GmvLo")
wtsStocks145GmvLo = round(wtsStocks145GmvLo,5)  
                                                     
# fit a fundamental factor model
fit.cross <- fitFfm(data = dat, 
              exposure.vars = c("SECTOR","ROE","BP","MOM121","SIZE","VOL121",
              "EP"),date.var = "DATE", ret.var = "RETURN", asset.var = "TICKER", 
              fit.method="WLS", z.score = TRUE)
repRisk(fit.cross, risk = "Sd", decomp = 'FCR', nrowPrint = 10,
        digits = 4) 
# get the factor contributions of risk 
repRisk(fit.cross, wtsStocks145GmvLo, risk = "Sd", decomp = 'FPCR', 
        nrowPrint = 10)               
# plot
repRisk(fit.cross, wtsStocks145GmvLo, risk = "Sd", decomp = 'FPCR', 
        isPrint = FALSE, isPlot = TRUE)  

AvinashAcharya/Test_factorAnalytics documentation built on May 5, 2019, 9:22 a.m.