Package 'PeerPerformance' reference manual

Title:	Luck-Corrected Peer Performance Analysis in R
Description:	Provides functions to perform the peer performance analysis of funds' returns as described in Ardia and Boudt (2018) <doi:10.1016/j.jbankfin.2017.10.014>.
Authors:	David Ardia [aut, cre, cph, fnd] (ORCID: <https://orcid.org/0000-0003-2823-782X>), Nabil Bouamara [ctb], Kris Boudt [aut] (ORCID: <https://orcid.org/0000-0002-1000-5142>), Sebastien Legros [ctb], Murilo Andre Peres Pereira [ctb] (ORCID: <https://orcid.org/0009-0007-0086-4801>), Benjamin Seguin [ctb]
Maintainer:	David Ardia <[email protected]>
License:	GPL (>= 2)
Version:	2.4.0
Built:	2026-06-25 11:43:28 UTC
Source:	https://github.com/ardiad/peerperformance

Screening using the alpha outperformance ratio

Description

Function which performs the screening of a universe of returns, and computes the alpha outperformance ratio.

Usage

alphaScreening(
  X,
  factors = NULL,
  control = list(),
  screen_beta = NULL,
  Y = NULL
)
alphaScreening(
  X,
  factors = NULL,
  control = list(),
  screen_beta = NULL,
  Y = NULL
)

Arguments

X

Matrix $(T \times N)$ of $T$ returns for the $N$ funds. NA values are allowed.

factors

Matrix $(T \times K)$ of $T$ returns for the $K$ factors. NA values are allowed.

control

Control parameters (see *Details*).

screen_beta

Boolean to screen all factors' coefficients (beta). Default: screen_beta = NULL, in which case the value is taken from control$screen_beta (itself defaulting to FALSE, i.e. only the alpha is screened). When supplied directly, the argument takes precedence over the control list. If TRUE, each element of the returned list will have a new first dimension representing each coefficient (the first one being alpha).

Y

Optional matrix $(T \times M)$ of returns for a second (peer) group of $M$ funds. When supplied, the ratios are computed for each fund in X against the funds in Y (cross-group screening) instead of against the other funds in X. A single focal fund versus a peer group corresponds to X being a vector (or a one-column matrix). Columns of Y identical to the focal fund (e.g. when X is a subset of Y) are automatically excluded. Default: Y = NULL, i.e. within-group screening.

Details

The alpha measure (Treynor and Black 1973, Carhart 1997, Fung and Hsieh 2004) is one industry standard for measuring the absolute risk adjusted performance of hedge funds. We propose to complement the alpha measure with the fund's alpha outperformance ratio, defined as the percentage number of funds that have a significantly lower alpha. In a pairwise testing framework, a fund can have a significantly higher alpha because of luck. We correct for this by applying the false discovery rate approach by Storey (2002).

The methodology proceeds as follows:

(1) compute all pairwise tests of alpha differences. This means that for a universe of $N$ funds, we perform $N(N-1)/2$ tests. The algorithm has been parallelized and the computational burden can be split across several cores. The number of cores can be defined in control, see below.
(2) for each fund, the false discovery rate approach by Storey (2002) is used to determine the proportions of over, equal, and underperforming funds, in terms of alpha, in the database.

The argument control is a list that can supply any of the following components:

'hac' Heteroscedastic-autocorrelation consistent standard errors. Default: hac = FALSE.
'minObs' Minimum number of concordant observations to compute the ratios. Default: minObs = 10.
'minObsPi' Minimum number of observations for computing the p-values). Default: minObsPi = 1.
'nCore' Number of cores used to perform the screening. Default: nCore = 1.
'lambda' Threshold value to compute pi0. Default: lambda = NULL, i.e. data driven choice.
'gammaPos' One-sided quantile level (of the standard Normal distribution) used as the critical value for counting outperformed peers. Default: gammaPos = 0.4 (the value recommended in Ardia and Boudt, 2018).
'gammaNeg' One-sided quantile level (of the standard Normal distribution) used as the critical value for counting peers that outperform the focal fund. Default: gammaNeg = 0.6.
'screen_beta' Screen the factor exposures (betas) in addition to the alpha; see the screen_beta argument. Default: screen_beta = FALSE.

Value

A list with the following components:

n: Vector (of length $N$ ) of number of non-NA observations.

npeer: Vector (of length $N$ ) of number of available peers.

alpha: Vector (of length $N$ ) of unconditional alpha.

dalpha: Matrix (of size $N \times N$ ) of alpha differences.

tstat: Matrix (of size $N \times N$ ) of t-statistics.

pval: Matrix (of size $N \times N$ ) of p-values of test for alpha differences.

lambda: Vector (of length $N$ ) of lambda values.

pizero: Vector (of length $N$ ) of probability of equal performance.

pipos: Vector (of length $N$ ) of probability of outperformance performance.

pineg: Vector (of length $N$ ) of probability of underperformance performance.

Note

Further details on the methodology with an application to the hedge fund industry is given in Ardia and Boudt (2018).

Application of the false discovery rate approach applied to the mutual fund industry has been presented in Barras, Scaillet and Wermers (2010).

Currently, the HAC asymptotic and studentized circular block bootstrap presented in Ledoit and Wolf (2008) are not supported by the alphaScreening function.

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Boudt, K. (2015). Testing equality of modified Sharpe ratios. Finance Research Letters 13, pp.97–104. doi:10.1016/j.frl.2015.02.008

Ardia, D., Boudt, K. (2018). The peer performance ratios of hedge funds. Journal of Banking and Finance 87, pp.351–368. doi:10.1016/j.jbankfin.2017.10.014

Barras, L., Scaillet, O., Wermers, R. (2010). False discoveries in mutual fund performance: Measuring luck in estimated alphas. Journal of Finance 65(1), pp.179–216.

Carhart, M. (1997). On persistence in mutual fund performance. Journal of Finance 52(1), pp.57–82.

Fama, E., French, K. (2010). Luck versus skill in the cross-section of mutual fund returns. Journal of Finance 65(5), pp.1915–1947.

Fung, W., Hsieh, D. (2004). Hedge fund benchmarks: A risk based approach. Financial Analysts Journal 60(5), pp.65–80.

Storey, J. (2002). A direct approach to false discovery rates. Journal of the Royal Statistical Society B 64(3), pp.479–498.

Treynor, J. L., Black, F. (1973). How to use security analysis to improve portfolio selection. Journal of Business 46(1), pp.66–86.

Examples

## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
rets = hfdata[,1:4]

## Run alpha screening
ctr = list(nCore = 1)
alphaScreening(rets, control = ctr)

## Run alpha screening with HAC standard deviation
ctr = list(nCore = 1, hac = TRUE)
alphaScreening(rets, control = ctr)

## Cross-group screening: a single focal fund against a peer group
alphaScreening(hfdata[, 1], Y = hfdata[, 11:20], control = list(nCore = 1))

## Cross-group screening: peer group X against peer group Y
alphaScreening(hfdata[, 1:5], Y = hfdata[, 11:20], control = list(nCore = 1))
## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
rets = hfdata[,1:4]

## Run alpha screening
ctr = list(nCore = 1)
alphaScreening(rets, control = ctr)

## Run alpha screening with HAC standard deviation
ctr = list(nCore = 1, hac = TRUE)
alphaScreening(rets, control = ctr)

## Cross-group screening: a single focal fund against a peer group
alphaScreening(hfdata[, 1], Y = hfdata[, 11:20], control = list(nCore = 1))

## Cross-group screening: peer group X against peer group Y
alphaScreening(hfdata[, 1:5], Y = hfdata[, 11:20], control = list(nCore = 1))

Testing the difference of alpha outperformance ratios

Description

Function which performs the testing of the difference of alpha outperformance ratios.

Usage

alphaTesting(x, y, factors = NULL, control = list(), screen_beta = NULL)
alphaTesting(x, y, factors = NULL, control = list(), screen_beta = NULL)

Arguments

x

Vector (of length $T$ ) of returns for the first fund. NA values are allowed.

y

Vector (of length $T$ ) returns for the second fund. NA values are allowed.

factors

Matrix $(T \times K)$ of $T$ returns for the $K$ factors. NA values are allowed.

control

Control parameters (see *Details*).

screen_beta

Boolean to screen all factors' coefficients (beta). Default: screen_beta = NULL, in which case the value is taken from control$screen_beta (itself defaulting to FALSE). When supplied directly, the argument takes precedence. screen_beta = TRUE requires factors.

Details

The alpha measure (Treynor and Black 1973, Carhart 1997, Fung and Hsieh 2004) is one industry standard for measuring the absolute risk adjusted performance of hedge funds. This function performs the testing of alpha outperformance ratio difference for two funds.

For the testing, only the intersection of non-NA observations for the two funds are used.

The argument control is a list that can supply any of the following components:

'hac' Heteroscedastic-autocorrelation consistent standard errors. Default: hac = FALSE.

Value

A list with the following components:

n: Number of non-NA concordant observations.

alpha: Vector (of length 2) of the two funds' unconditional alphas.

dalpha: Difference of the two funds' alphas.

tstat: t-stat of the alpha difference.

pval: p-value of the test of equal alpha.

When screen_beta = TRUE, alpha is a matrix (one row per coefficient, the first being the alpha) and dalpha, tstat and pval are vectors with one entry per coefficient; the print method reports the alpha (first) coefficient only.

Note

Further details on the methodology with an application to the hedge fund industry is given in Ardia and Boudt (2018).

Some internal functions where adapted from Michael Wolf MATLAB code.

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Boudt, K. (2015). Testing equality of modified Sharpe ratios. Finance Research Letters 13, 97–104.

Ardia, D., Boudt, K. (2018). The peer performance ratios of hedge funds. Journal of Banking and Finance 87, 351–368.

Barras, L., Scaillet, O., Wermers, R. (2010). False discoveries in mutual fund performance: Measuring luck in estimated alphas. Journal of Finance 65(1), 179–216.

Sharpe, W.F. (1994). The Sharpe ratio. Journal of Portfolio Management 21(1), 49–58.

Ledoit, O., Wolf, M. (2008). Robust performance hypothesis testing with the Sharpe ratio. Journal of Empirical Finance 15(5), 850–859.

Storey, J. (2002). A direct approach to false discovery rates. Journal of the Royal Statistical Society B 64(3), 479–498.

Examples

## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
x = hfdata[,1]
y = hfdata[,2]

## Run alpha testing
alphaTesting(x, y)
## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
x = hfdata[,1]
y = hfdata[,2]

## Run alpha testing
alphaTesting(x, y)

Coerce a 'SCREENING' object to a data frame

Description

Returns a tidy data.frame with one row per fund (or, when the screening was run with screen_beta = TRUE, one row per fund/coefficient), suitable for downstream filtering, ranking, and plotting.

Usage

## S3 method for class 'SCREENING'
as.data.frame(x, row.names = NULL, optional = FALSE, ...)
## S3 method for class 'SCREENING'
as.data.frame(x, row.names = NULL, optional = FALSE, ...)

Arguments

x

A SCREENING object.

row.names, optional

Currently ignored.

...

Further arguments (ignored).

Value

A data.frame with columns fund, the performance measure (alpha, sharpe, or msharpe), pipos, pizero, pineg, npeer, and lambda; a coefficient column is added when screen_beta = TRUE.

Author(s)

David Ardia and Kris Boudt.

Examples


  data("hfdata")
  sc <- alphaScreening(hfdata[, 1:10], control = list(nCore = 1))
  head(as.data.frame(sc))

data("hfdata")
  sc <- alphaScreening(hfdata[, 1:10], control = list(nCore = 1))
  head(as.data.frame(sc))

Bootstrap confidence intervals for the peer performance ratios

Description

Computes confidence intervals for one of the peer performance ratios ( $\hat\pi^+$ , $\hat\pi^0$ , or $\hat\pi^-$ ) of a SCREENING object, by a nonparametric peer (pairwise) bootstrap: for each focal fund its peers are resampled with replacement and the ratio is recomputed from the stored pairwise p-values and test statistics. The interval therefore reflects the peer-composition and aggregation uncertainty, holding the individual pairwise test statistics fixed.

Usage

## S3 method for class 'SCREENING'
confint(
  object,
  parm = c("pipos", "pizero", "pineg"),
  level = 0.95,
  ...,
  nBoot = 499L,
  gammaPos = 0.4,
  gammaNeg = 0.6
)
## S3 method for class 'SCREENING'
confint(
  object,
  parm = c("pipos", "pizero", "pineg"),
  level = 0.95,
  ...,
  nBoot = 499L,
  gammaPos = 0.4,
  gammaNeg = 0.6
)

Arguments

object

A SCREENING object (within-group or cross-group; not screen_beta).

parm

Which ratio to bound: "pipos" (default), "pizero", or "pineg".

level

Confidence level. Default: 0.95.

...

Unused.

nBoot

Number of bootstrap resamples. Default: 499.

gammaPos, gammaNeg

One-sided thresholds used to recompute the ratios (should match those of the original screening). Defaults: 0.4, 0.6.

Value

A matrix with one row per fund and two columns (lower and upper bounds); the point estimate is attached as attribute "estimate".

Author(s)

David Ardia and Kris Boudt.

Examples


data("hfdata")
set.seed(1234)
sc <- alphaScreening(hfdata[, 1:15], control = list(nCore = 1))
confint(sc, parm = "pipos")

data("hfdata")
set.seed(1234)
sc <- alphaScreening(hfdata[, 1:15], control = list(nCore = 1))
confint(sc, parm = "pipos")

Factor exposure heterogeneity from a beta screening

Description

Aggregates the per-fund equal-performance ratios of a screen_beta = TRUE alphaScreening into the factor-by-factor heterogeneity measure of Ardia et al. (2023). For each coefficient $k$ (the alpha and each factor beta) it reports the average equal-exposure ratio $\pi^0_k = \frac1N\sum_i \pi^0_{i,k}$ and the heterogeneity $1-\pi^0_k$ : the share of peers that are significantly differentiated on coefficient $k$ . A value close to one indicates large heterogeneity (much room to differentiate); close to zero, homogeneity.

Usage

exposureHeterogeneity(object)
exposureHeterogeneity(object)

Arguments

object

A SCREENING object produced with screen_beta = TRUE.

Value

A data.frame of class exposureHeterogeneity with columns coefficient, equalExposure ( $\pi^0_k$ ), and heterogeneity ( $1-\pi^0_k$ ).

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Bluteau, K., Lortie-Cloutier, G., Tran, D. (2023). Factor exposure heterogeneity in green and brown stocks. Finance Research Letters 55, Part A, 103900.

Examples


data("hfdata")
set.seed(1234)
fac <- matrix(rnorm(nrow(hfdata) * 2), ncol = 2,
              dimnames = list(NULL, c("MKT", "SMB")))
sc <- alphaScreening(hfdata[, 1:20], factors = fac, screen_beta = TRUE,
                     control = list(nCore = 1))
exposureHeterogeneity(sc)

data("hfdata")
set.seed(1234)
fac <- matrix(rnorm(nrow(hfdata) * 2), ncol = 2,
              dimnames = list(NULL, c("MKT", "SMB")))
sc <- alphaScreening(hfdata[, 1:20], factors = fac, screen_beta = TRUE,
                     control = list(nCore = 1))
exposureHeterogeneity(sc)

Hedge fund data

Description

The matrix hfdata contains randomized and modified monthly returns of hedge funds.

Usage

data(hfdata)
data(hfdata)

Format

A matrix of size 60 x 100, containing 60 monthly net returns for 100 hedge funds. Returns have been modified.

Source

HFR database

Compute modified Sharpe ratio

Description

Function which computes the modified Sharpe ratio

Usage

msharpe(X, level = 0.9, na.rm = TRUE, na.neg = TRUE)
msharpe(X, level = 0.9, na.rm = TRUE, na.neg = TRUE)

Arguments

X

Vector (of length $T$ ) or matrix (of size $T \times N$ ) of returns. NA values are allowed.

level

Modified Value-at-Risk level. Default: level = 0.90.

na.rm

A logical value indicating whether NA values should be stripped before the computation. Default na.rm = TRUE.

na.neg

A logical value indicating whether NA values should be returned if a negative modified Value-at-Risk is obtained. Default na.neg = TRUE.

Details

The modified Sharpe ratio (Favre and Galeano 2002) is one industry standard for measuring the absolute risk adjusted performance of hedge funds.

Value

Scalar or a vector (of size $N$ ) with the modified Sharpe ratios.

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Boudt, K. (2015). Testing equality of modified Sharpe ratios. Finance Research Letters 13, 97–104.

Ardia, D., Boudt, K. (2018). The peer performance ratios of hedge Funds. Journal of Banking and Finance 87, 351–368.

Favre, L., Galeano, J.A. (2002). Mean-modified Value-at-Risk Optimization with Hedge Funds. Journal of Alternative Investments 5(2), 21–25.

Gregoriou, G. N., Gueyie, J.-P. (2003). Risk-adjusted performance of funds of hedge funds using a modified Sharpe ratio. Journal of Wealth Management 6(3), 77–83.

Examples

## Load the data (randomized data of monthly hedge fund returns)
data('hfdata')

out = msharpe(hfdata)
print(out)

out = msharpe(hfdata, na.rm = FALSE)
print(out)
## Load the data (randomized data of monthly hedge fund returns)
data('hfdata')

out = msharpe(hfdata)
print(out)

out = msharpe(hfdata, na.rm = FALSE)
print(out)

Screening using the modified Sharpe outperformance ratio

Description

Function which performs the screening of a universe of returns, and computes the modified Sharpe outperformance ratio.

Usage

msharpeScreening(X, level = 0.9, na.neg = TRUE, control = list(), Y = NULL)
msharpeScreening(X, level = 0.9, na.neg = TRUE, control = list(), Y = NULL)

Arguments

X

Matrix $(T \times N)$ of $T$ returns for the $N$ funds. NA values are allowed.

level

Modified Value-at-Risk level. Default: level = 0.90.

na.neg

A logical value indicating whether NA values should be returned if a negative modified Value-at-Risk is obtained. Default na.neg = TRUE.

control

Control parameters (see *Details*).

Y

Optional matrix $(T \times M)$ of returns for a second (peer) group of $M$ funds. When supplied, the ratios are computed for each fund in X against the funds in Y (cross-group screening) instead of against the other funds in X; a single focal fund versus a peer group corresponds to X being a vector. Columns of Y identical to the focal fund are automatically excluded. Default: Y = NULL, i.e. within-group screening.

Details

The modified Sharpe ratio (Favre and Galeano 2002, Gregoriou and Gueyie 2003) is one industry standard for measuring the absolute risk adjusted performance of hedge funds. We propose to complement the modified Sharpe ratio with the fund's outperformance ratio, defined as the percentage number of funds that have a significantly lower modified Sharpe ratio. In a pairwise testing framework, a fund can have a significantly higher modified Sharpe ratio because of luck. We correct for this by applying the false discovery rate approach by Storey (2002).

For the testing, only the intersection of non-NA observations for the two funds are used.

The argument control is a list that can supply any of the following components:

'type' Asymptotic approach (type = 1) or studentized circular bootstrap approach (type = 2). Default: type = 1.
'ttype' Test based on ratio (type = 1) or product (type = 2). Default: type = 2.
'hac' heteroscedastic-autocorrelation consistent standard errors. Default: hac = FALSE.
'nBoot' Number of bootstrap replications for computing the p-value. Default: nBoot = 499.
'bBoot' Block length in the circular bootstrap. Default: bBoot = 1, i.e. iid bootstrap. (The data-driven choice bBoot = 0 is only available in msharpeTesting, not in screening.)
'pBoot' Symmetric p-value (pBoot = 1) or asymmetric p-value (pBoot = 2). Default: pBoot = 1.
'nCore' Number of cores to be used. Default: nCore = 1.
'minObs' Minimum number of concordant observations to compute the ratios. Default: minObs = 10.
'minObsPi' Minimum number of observations to compute pi0. Default: minObsPi = 1.
'lambda' Threshold value to compute pi0. Default: lambda = NULL, i.e. data driven choice.
'gammaPos' One-sided quantile level (of the standard Normal distribution) used as the critical value for counting outperformed peers. Default: gammaPos = 0.4 (the value recommended in Ardia and Boudt, 2018).
'gammaNeg' One-sided quantile level (of the standard Normal distribution) used as the critical value for counting peers that outperform the focal fund. Default: gammaNeg = 0.6.

Value

A list with the following components:

n: Vector (of length $N$ ) of number of non-NA observations.

npeer: Vector (of length $N$ ) of number of available peers.

msharpe: Vector (of length $N$ ) of unconditional modified Sharpe ratios.

dmsharpe: Matrix (of size $N \times N$ ) of modified Sharpe ratios differences.

tstat: Matrix (of size $N \times N$ ) of t-statistics.

pval: Matrix (of size $N \times N$ ) of p-values of test for modified Sharpe ratios differences.

lambda: Vector (of length $N$ ) of lambda values.

pizero: Vector (of length $N$ ) of probability of equal performance.

pipos: Vector (of length $N$ ) of probability of outperformance performance.

pineg: Vector (of length $N$ ) of probability of underperformance performance.

Note

Further details on the methodology with an application to the hedge fund industry is given in Ardia and Boudt (2018).

Some internal functions where adapted from Michael Wolf MATLAB code.

Application of the false discovery rate approach applied to the mutual fund industry has been presented in Barras, Scaillet and Wermers (2010).

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Boudt, K. (2015). Testing equality of modified Sharpe ratios. Finance Research Letters 13, 97–104.

Ardia, D., Boudt, K. (2018). The peer performance ratios of hedge funds. Journal of Banking and Finance 87, 351–368.

Barras, L., Scaillet, O., Wermers, R. (2010). False discoveries in mutual fund performance: Measuring luck in estimated alphas. Journal of Finance 65(1), 179–216.

Favre, L., Galeano, J.A. (2002). Mean-modified Value-at-Risk Optimization with Hedge Funds. Journal of Alternative Investments 5(2), 21–25.

Gregoriou, G. N., Gueyie, J.-P. (2003). Risk-adjusted performance of funds of hedge funds using a modified Sharpe ratio. Journal of Wealth Management 6(3), 77–83.

Ledoit, O., Wolf, M. (2008). Robust performance hypothesis testing with the Sharpe ratio. Journal of Empirical Finance 15(5), 850–859.

Storey, J. (2002). A direct approach to false discovery rates. Journal of the Royal Statistical Society B 64(3), 479–498.

Examples

## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
rets = hfdata[,1:4]

## Modified Sharpe screening
msharpeScreening(rets, control = list(nCore = 1))

## Modified Sharpe screening with bootstrap and HAC standard deviation
msharpeScreening(rets, control = list(nCore = 1, type = 2, hac = TRUE))
## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
rets = hfdata[,1:4]

## Modified Sharpe screening
msharpeScreening(rets, control = list(nCore = 1))

## Modified Sharpe screening with bootstrap and HAC standard deviation
msharpeScreening(rets, control = list(nCore = 1, type = 2, hac = TRUE))

Testing the difference of modified Sharpe ratios

Description

Function which performs the testing of the difference of modified Sharpe ratios.

Usage

msharpeTesting(x, y, level = 0.9, na.neg = TRUE, control = list())
msharpeTesting(x, y, level = 0.9, na.neg = TRUE, control = list())

Arguments

x

Vector (of length $T$ ) of returns for the first fund. NA values are allowed.

y

Vector (of length $T$ ) of returns for the second fund. NA values are allowed.

level

Modified Value-at-Risk level. Default: level = 0.90.

na.neg

A logical value indicating whether NA values should be returned if a negative modified Value-at-Risk is obtained. Default na.neg = TRUE.

control

Control parameters (see *Details*).

Details

The modified Sharpe ratio (Favre and Galeano 2002) is one industry standard for measuring the absolute risk adjusted performance of hedge funds. This function performs the testing of modified Sharpe ratio difference for two funds using a similar approach than Ledoit and Wolf (2002). See also Gregoriou and Gueyie (2003).

For the testing, only the intersection of non-NA observations for the two funds are used.

The argument control is a list that can supply any of the following components:

'type' Asymptotic approach (type = 1) or studentized circular bootstrap approach (type = 2). Default: type = 1.
'ttype' Test based on ratio (type = 1) or product (type = 2). Default: type = 2.
'hac' Heteroscedastic-autocorrelation consistent standard errors. Default: hac = FALSE.
'minObs' Minimum number of concordant observations to compute the ratios. Default: minObs = 10.
'nBoot' Number of bootstrap replications for computing the p-value. Default: nBoot = 499.
'bBoot' Block length in the circular bootstrap. Default: bBoot = 1, i.e. iid bootstrap. bBoot = 0 uses optimal block-length.
'pBoot' Symmetric p-value (pBoot = 1) or asymmetric p-value (pBoot = 2). Default: pBoot = 1.

Value

A list with the following components:

n: Number of non-NA concordant observations.

msharpe: Vector (of length 2) of unconditional modified Sharpe ratios.

dmsharpe: Modified Sharpe ratios difference.

tstat: t-stat of modified Sharpe ratios differences.

pval: pvalues of test of modified Sharpe ratios differences.

Note

Further details on the methodology with an application to the hedge fund industry is given in Ardia and Boudt (2018).

Some internal functions where adapted from Michael Wolf MATLAB code.

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Boudt, K. (2015). Testing equality of modified Sharpe ratios. Finance Research Letters 13, pp.97–104. doi:10.1016/j.frl.2015.02.008

Ardia, D., Boudt, K. (2018). The peer performance ratios of hedge funds. Journal of Banking and Finance 87, pp.351–368. doi:10.1016/j.jbankfin.2017.10.014

Barras, L., Scaillet, O., Wermers, R. (2010). False discoveries in mutual fund performance: Measuring luck in estimated alphas. Journal of Finance 65(1), pp.179–216.

Favre, L., Galeano, J.A. (2002). Mean-modified Value-at-Risk Optimization with Hedge Funds. Journal of Alternative Investments 5(2), pp.21–25.

Gregoriou, G. N., Gueyie, J.-P. (2003). Risk-adjusted performance of funds of hedge funds using a modified Sharpe ratio. Journal of Wealth Management 6(3), pp.77–83.

Ledoit, O., Wolf, M. (2008). Robust performance hypothesis testing with the Sharpe ratio. Journal of Empirical Finance 15(5), pp.850–859.

Storey, J. (2002). A direct approach to false discovery rates. Journal of the Royal Statistical Society B 64(3), pp.479–498.

Examples

## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
x = hfdata[,1]
y = hfdata[,2]

## Run modified Sharpe testing (asymptotic)
ctr = list(type = 1)
out = msharpeTesting(x, y, level = 0.95, control = ctr)
print(out)

## Run modified Sharpe testing (asymptotic hac)
ctr = list(type = 1, hac = TRUE)
out = msharpeTesting(x, y, level = 0.95, control = ctr)
print(out)

## Run modified Sharpe testing (iid bootstrap)
set.seed(1234)
ctr = list(type = 2, nBoot = 250)
out = msharpeTesting(x, y, level = 0.95, control = ctr)
print(out)

## Run modified Sharpe testing (circular bootstrap)
set.seed(1234)
ctr = list(type = 2, nBoot = 250, bBoot = 5)
out = msharpeTesting(x, y, level = 0.95, control = ctr)
print(out)
## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
x = hfdata[,1]
y = hfdata[,2]

## Run modified Sharpe testing (asymptotic)
ctr = list(type = 1)
out = msharpeTesting(x, y, level = 0.95, control = ctr)
print(out)

## Run modified Sharpe testing (asymptotic hac)
ctr = list(type = 1, hac = TRUE)
out = msharpeTesting(x, y, level = 0.95, control = ctr)
print(out)

## Run modified Sharpe testing (iid bootstrap)
set.seed(1234)
ctr = list(type = 2, nBoot = 250)
out = msharpeTesting(x, y, level = 0.95, control = ctr)
print(out)

## Run modified Sharpe testing (circular bootstrap)
set.seed(1234)
ctr = list(type = 2, nBoot = 250, bBoot = 5)
out = msharpeTesting(x, y, level = 0.95, control = ctr)
print(out)

PeerPerformance: Luck-corrected peer performance analysis in R

Description

PeerPerformance is an R package for the peer-performance evaluation of financial investments with luck-correction, useful in the financial industry. In particular, it implements the peer performance ratios of Ardia and Boudt (2018) which measure the percentage of peers a focal (hedge) fund outperforms and underperforms, after correction for luck. It is useful for fund or portfolio managers to benchmark their investments or screen a universe of new funds. In addition, the package implements the testing framework for the Sharpe and modified Sharpe ratios, described in Ledoit and Wolf (2008) and Ardia and Boudt (2015). See also Ardia et al. (2022,2023) for applications in sustainable finance.

Functions

Alpha: alphaTesting and alphaScreening;
Sharpe ratio: sharpe, sharpeTesting and sharpeScreening;
Modified Sharpe ratio: msharpe, msharpeTesting and msharpeScreening;
Screening: alphaScreening, sharpeScreening, and msharpeScreening.

Update

The latest version of the package is available at https://github.com/ArdiaD/PeerPerformance

Note

By using PeerPerformance you agree to the following rules: (1) You must cite Ardia and Boudt (2018) in working papers and published papers that use PeerPerformance (use citation("PeerPerformance")), (2) you must place the URL https://CRAN.R-project.org/package=PeerPerformance in a footnote to help others find PeerPerformance, and (3) you assume all risk for the use of PeerPerformance.

Full description of the methodologies implemented in the various functions is available in Ledoit and Wolf (2008) and Ardia and Boudt (2015, 2018).

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Boudt, K. (2015). Testing equality of modified Sharpe ratios. Finance Research Letters 13, 97–104.

Ardia, D., Boudt, K. (2018). The peer performance ratios of hedge Funds. Journal of Banking and Finance 87, 351–368.

Ardia, D., Bluteau, K., Tran, D. (2022). How easy is it for investment managers to deploy their talent in green and brown stocks? Finance Research Letters 48, 102992.

Ardia, D., Bluteau, K., Lortie-Cloutier, G., Tran, D. (2023). Factor exposure heterogeneity in green and brown stocks. Finance Research Letters 55, Part A, 103900.

Barras, L., Scaillet, O., Wermers, R. (2010). False discoveries in mutual fund performance: Measuring luck in estimated alphas. Journal of Finance 65(1), 179–216.

Favre, L., Galeano, J.A. (2002). Mean-modified Value-at-Risk Optimization with Hedge Funds. Journal of Alternative Investments 5(2), 21–25.

Gregoriou, G. N., Gueyie, J.-P. (2003). Risk-adjusted performance of funds of hedge funds using a modified Sharpe ratio. Journal of Wealth Management 6(3), 77–83.

Ledoit, O., Wolf, M. (2008). Robust performance hypothesis testing with the Sharpe ratio. Journal of Empirical Finance 15(5), 850–859.

Sharpe, W.F. (1994). The Sharpe ratio. Journal of Portfolio Management 21(1), 49–58.

Storey, J. (2002). A direct approach to false discovery rates. Journal of the Royal Statistical Society B 64(3), 479–498.

Plot factor exposure heterogeneity

Description

Bar plot of the heterogeneity measure $1-\pi^0_k$ returned by exposureHeterogeneity.

Usage

## S3 method for class 'exposureHeterogeneity'
plot(x, ...)
## S3 method for class 'exposureHeterogeneity'
plot(x, ...)

Arguments

x

An exposureHeterogeneity object.

...

Further arguments passed to barplot.

Value

Invisibly returns x.

Author(s)

David Ardia and Kris Boudt.

Examples


  data("hfdata")
  sc <- alphaScreening(hfdata[, 1:10], factors = hfdata[, 50:51],
                       control = list(nCore = 1, screen_beta = TRUE))
  plot(exposureHeterogeneity(sc))

data("hfdata")
  sc <- alphaScreening(hfdata[, 1:10], factors = hfdata[, 50:51],
                       control = list(nCore = 1, screen_beta = TRUE))
  plot(exposureHeterogeneity(sc))

Plot a rolling peer performance screening

Description

Time-series plot of the rolling peer performance ratios produced by rollScreening.

Usage

## S3 method for class 'rollScreening'
plot(x, what = NULL, legend = TRUE, ...)
## S3 method for class 'rollScreening'
plot(x, what = NULL, legend = TRUE, ...)

Arguments

x

A rollScreening object.

what

Either "heterogeneity" ( $1-\bar\pi^0$ ) or "ratios" (the averaged $\bar\pi^+,\bar\pi^0,\bar\pi^-$ ). Default: "heterogeneity" for a beta screening, "ratios" otherwise.

legend

A logical value indicating whether to draw a legend. Default: TRUE.

...

Further graphical arguments.

Value

Invisibly returns x.

Author(s)

David Ardia and Kris Boudt.

Examples


  data("hfdata")
  roll <- rollScreening(hfdata[, 1:15], width = 36, by = 12,
                        control = list(nCore = 1))
  plot(roll)

data("hfdata")
  roll <- rollScreening(hfdata[, 1:15], width = 36, by = 12,
                        control = list(nCore = 1))
  plot(roll)

Peer performance screening plot

Description

Plot method for the SCREENING object returned by alphaScreening, sharpeScreening and msharpeScreening. It reproduces the peer performance screening plot of Ardia and Boudt (2018): funds are sorted by their performance measure and, for each fund, the estimated outperformance ( $\hat\pi^+$ ), equal-performance ( $\hat\pi^0$ ) and underperformance ( $\hat\pi^-$ ) ratios are displayed as a horizontal stacked bar. The dashed diagonal lines depict the naive percentile-rank benchmark ( $\hat\pi^0 = 0$ ); the gap between the black $\hat\pi^+$ area and the diagonal visualizes the luck correction.

Usage

## S3 method for class 'SCREENING'
plot(
  x,
  nblock = NULL,
  reference = NULL,
  band = 27.5,
  colorset = c(grDevices::gray(0), grDevices::gray(0.8), grDevices::gray(0.5)),
  ...
)
## S3 method for class 'SCREENING'
plot(
  x,
  nblock = NULL,
  reference = NULL,
  band = 27.5,
  colorset = c(grDevices::gray(0), grDevices::gray(0.8), grDevices::gray(0.5)),
  ...
)

Arguments

x

A SCREENING object.

nblock

Optional number of equally-sized blocks into which the funds (sorted by performance) are aggregated by averaging, as in Ardia and Boudt (2018). Default: nblock = NULL, i.e. one bar per fund.

reference

A logical value indicating whether the dashed percentile-rank reference lines should be drawn. Default: reference = NULL, i.e. drawn for within-group screening and omitted for cross-group (Y-based) screening, where the within-group percentile-rank benchmark does not apply. A single focal fund (cross-group with one fund in X) is shown as a single stacked bar.

band

Half-width (in percentage points) of the reference band drawn around the central diagonal. Default: band = 27.5.

colorset

Vector of three colors for the $\hat\pi^+$ , $\hat\pi^0$ and $\hat\pi^-$ areas. Default: c(gray(0), gray(0.8), gray(0.5)).

...

Further graphical arguments passed to barplot.

Details

If the screening was run with screen_beta = TRUE, the alpha (first) coefficient is used.

Value

Invisibly returns the (sorted, possibly aggregated) matrix of ratios that is plotted.

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Boudt, K. (2018). The peer performance ratios of hedge funds. Journal of Banking and Finance 87, pp.351–368. doi:10.1016/j.jbankfin.2017.10.014

Examples


data("hfdata")
set.seed(1234)
sc <- alphaScreening(hfdata[, 1:30], control = list(nCore = 1))
plot(sc)

data("hfdata")
set.seed(1234)
sc <- alphaScreening(hfdata[, 1:30], control = list(nCore = 1))
plot(sc)

Print method for the 'SCREENING' object

Description

Print method for the SCREENING object returned by alphaScreening, sharpeScreening and msharpeScreening.

Usage

## S3 method for class 'SCREENING'
print(x, ...)
## S3 method for class 'SCREENING'
print(x, ...)

Arguments

x

A SCREENING object.

...

Further arguments passed to or from other methods.

Value

Invisibly returns x.

Author(s)

David Ardia and Kris Boudt.

Print method for the 'summary.SCREENING' object

Description

Print method for the object returned by summary.SCREENING.

Usage

## S3 method for class 'summary.SCREENING'
print(x, ...)
## S3 method for class 'summary.SCREENING'
print(x, ...)

Arguments

x

A summary.SCREENING object.

...

Further arguments (ignored).

Value

Invisibly returns x.

Author(s)

Murilo Andre Peres Pereira, David Ardia and Kris Boudt.

Print method for the 'TESTING' object

Description

Print method for the TESTING object returned by sharpeTesting, msharpeTesting and alphaTesting.

Usage

## S3 method for class 'TESTING'
print(x, ...)
## S3 method for class 'TESTING'
print(x, ...)

Arguments

x

A TESTING object.

...

Further arguments passed to or from other methods.

Value

Invisibly returns x.

Author(s)

David Ardia and Kris Boudt.

Rolling-window peer performance screening

Description

Runs a peer performance screening on a rolling time window and returns the time series of cross-sectionally averaged peer performance ratios. This is the design underlying the dynamic analyses of Ardia et al. (2022, 2023): on each window the screening is computed and the ratios are averaged across funds (per coefficient when screen_beta = TRUE), yielding a series of (factor exposure) heterogeneity measures.

Usage

rollScreening(
  X,
  factors = NULL,
  Y = NULL,
  screen = c("alpha", "sharpe", "msharpe"),
  width = 36L,
  by = 1L,
  level = 0.9,
  na.neg = TRUE,
  control = list(),
  dates = NULL
)
rollScreening(
  X,
  factors = NULL,
  Y = NULL,
  screen = c("alpha", "sharpe", "msharpe"),
  width = 36L,
  by = 1L,
  level = 0.9,
  na.neg = TRUE,
  control = list(),
  dates = NULL
)

Arguments

X

Matrix $(T \times N)$ of $T$ returns for the $N$ funds.

factors

Matrix $(T \times K)$ of factor returns (for screen = "alpha"). Default: NULL.

Y

Optional matrix $(T \times M)$ of returns for a peer group (cross-group screening, see alphaScreening). Default: NULL.

screen

Performance measure to screen on: "alpha" (default), "sharpe", or "msharpe".

width

Window length (number of observations). Default: 36.

by

Step (in observations) between successive windows. Default: 1.

level

Modified Value-at-Risk level (for screen = "msharpe"). Default: 0.9.

na.neg

A logical value passed to msharpeScreening (for screen = "msharpe") indicating whether a negative modified Value-at-Risk yields NA. Default: TRUE.

control

Control parameters passed to the screening function (see alphaScreening); set control = list(screen_beta = TRUE) to obtain per-factor heterogeneity series.

dates

Optional vector of length $T$ (e.g. Date) used to label the windows by their end date. Default: NULL.

Value

A data.frame of class rollScreening with one row per window (or per window/coefficient when screen_beta = TRUE), with columns window, index (end-of-window row), optionally date, coefficient, the averaged pizero/pipos/ pineg, and heterogeneity ( $1-\bar\pi^0$ ).

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Bluteau, K., Tran, D. (2022). How easy is it for investment managers to deploy their talent in green and brown stocks? Finance Research Letters 48, 102992.

Ardia, D., Bluteau, K., Lortie-Cloutier, G., Tran, D. (2023). Factor exposure heterogeneity in green and brown stocks. Finance Research Letters 55, Part A, 103900.

Examples


data("hfdata")
set.seed(1234)
roll <- rollScreening(hfdata, screen = "alpha", width = 36, by = 6,
                      control = list(nCore = 1))
plot(roll)

data("hfdata")
set.seed(1234)
roll <- rollScreening(hfdata, screen = "alpha", width = 36, by = 6,
                      control = list(nCore = 1))
plot(roll)

Compute Sharpe ratio

Description

Function which computes the Sharpe ratio.

Usage

sharpe(X, na.rm = TRUE)
sharpe(X, na.rm = TRUE)

Arguments

X

Vector (of length $T$ ) or matrix (of size $T \times N$ ) of returns for $N$ funds. NA values are allowed.

na.rm

A logical value indicating whether NA values should be stripped before the computation. Default na.rm = TRUE

Details

The Sharpe ratio (Sharpe 1994) is one industry standard for measuring the absolute risk adjusted performance of hedge funds.

Value

A scalar or a vector (of size $N$ ) with the Sharpe ratios.

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Boudt, K. (2015). Testing equality of modified Sharpe ratios. Finance Research Letters 13, pp.97–104. doi:10.1016/j.frl.2015.02.008

Ardia, D., Boudt, K. (2016). The Peer Ratios Performance of Hedge Funds. Working paper. doi:10.2139/ssrn.2000901

Sharpe, W.F. (1994). The Sharpe ratio. Journal of Portfolio Management 21(1), pp.49–58. doi:10.3905/jpm.1994.409501

Examples

## Load the data
data('hfdata')

## Compute the Sharpe ratio
out = sharpe(hfdata)
print(out)

out = sharpe(hfdata, na.rm = FALSE)
print(out)
## Load the data
data('hfdata')

## Compute the Sharpe ratio
out = sharpe(hfdata)
print(out)

out = sharpe(hfdata, na.rm = FALSE)
print(out)

Screening using the Sharpe outperformance ratio

Description

Function which performs the screening of a universe of returns, and computes the Sharpe outperformance ratio.

Usage

sharpeScreening(X, control = list(), Y = NULL)
sharpeScreening(X, control = list(), Y = NULL)

Arguments

X

Matrix $(T \times N)$ of $T$ returns for the $N$ funds. NA values are allowed.

control

Control parameters (see *Details*).

Y

Details

The Sharpe ratio (Sharpe 1994) is one industry standard for measuring the absolute risk adjusted performance of hedge funds. We propose to complement the Sharpe ratio with the fund's outperformance ratio, defined as the percentage number of funds that have a significantly lower Sharpe ratio. In a pairwise testing framework, a fund can have a significantly higher Sharpe ratio because of luck. We correct for this by applying the false discovery rate approach by Storey (2002).

For the testing, only the intersection of non-NA observations for the two funds are used.

The methodology proceeds as follows:

(1) compute all pairwise tests of Sharpe differences using the bootstrap approach of Ledoit and Wolf (2002). This means that for a universe of $N$ funds, we perform $N(N-1)/2$ tests. The algorithm has been parallelized and the computational burden can be split across several cores. The number of cores can be defined in control, see below.
(2) for each fund, the false discovery rate approach by Storey (2002) is used to determine the proportions over, equal, and underperforming funds, in terms of Sharpe ratio, in the database.

The argument control is a list that can supply any of the following components:

'type' Asymptotic approach (type = 1) or studentized circular bootstrap approach (type = 2). Default: type = 1.
'ttype' Test based on ratio (type = 1) or product (type = 2). Default: type = 2.
'hac' Heteroscedastic-autocorrelation consistent standard errors. Default: hac = FALSE.
'nBoot' Number of bootstrap replications for computing the p-value. Default: nBoot = 499.
'bBoot' Block length in the circular bootstrap. Default: bBoot = 1, i.e. iid bootstrap. (The data-driven choice bBoot = 0 is only available in sharpeTesting, not in screening.)
'pBoot' Symmetric p-value (pBoot = 1) or asymmetric p-value (pBoot = 2). Default: pBoot = 1.
'nCore' Number of cores to be used. Default: nCore = 1.
'minObs' Minimum number of concordant observations to compute the ratios. Default: minObs = 10.
'minObsPi' Minimum number of observations to compute pi0. Default: minObsPi = 1.
'lambda' Threshold value to compute pi0. Default: lambda = NULL, i.e. data driven choice.
'gammaPos' One-sided quantile level (of the standard Normal distribution) used as the critical value for counting outperformed peers. Default: gammaPos = 0.4 (the value recommended in Ardia and Boudt, 2018).
'gammaNeg' One-sided quantile level (of the standard Normal distribution) used as the critical value for counting peers that outperform the focal fund. Default: gammaNeg = 0.6.

Value

A list with the following components:

n: Vector (of length $N$ ) of number of non-NA observations.

npeer: Vector (of length $N$ ) of number of available peers.

sharpe: Vector (of length $N$ ) of unconditional Sharpe ratios.

dsharpe: Matrix (of size $N \times N$ ) of Sharpe ratios differences.

tstat: Matrix (of size $N \times N$ ) of t-statistics.

pval: Matrix (of size $N \times N$ ) of pvalues of test for Sharpe ratios differences.

lambda: vector (of length $N$ ) of lambda values.

pizero: vector (of length $N$ ) of probability of equal performance.

pipos: vector (of length $N$ ) of probability of outperformance performance.

pineg: Vector (of length $N$ ) of probability of underperformance performance.

Note

Further details on the methodology with an application to the hedge fund industry is given in Ardia and Boudt (2018).

Some internal functions where adapted from Michael Wolf MATLAB code.

Application of the false discovery rate approach applied to the mutual fund industry has been presented in Barras, Scaillet and Wermers (2010).

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Boudt, K. (2015). Testing equality of modified Sharpe ratios. Finance Research Letters 13, 97–104.

Ardia, D., Boudt, K. (2018). The peer performance ratios of hedge funds. Journal of Banking and Finance 87, 351–368.

Barras, L., Scaillet, O., Wermers, R. (2010). False discoveries in mutual fund performance: Measuring luck in estimated alphas. Journal of Finance 65(1), 179–216.

Sharpe, W.F. (1994). The Sharpe ratio. Journal of Portfolio Management 21(1), 49–58.

Ledoit, O., Wolf, M. (2008). Robust performance hypothesis testing with the Sharpe ratio. Journal of Empirical Finance 15(5), 850–859.

Storey, J. (2002). A direct approach to false discovery rates. Journal of the Royal Statistical Society B 64(3), 479–498.

Examples

## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
rets = hfdata[,1:4]

## Sharpe screening
sharpeScreening(rets, control = list(nCore = 1))

## Sharpe screening with bootstrap and HAC standard deviation
sharpeScreening(rets, control = list(nCore = 1, type = 2, hac = TRUE))
## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
rets = hfdata[,1:4]

## Sharpe screening
sharpeScreening(rets, control = list(nCore = 1))

## Sharpe screening with bootstrap and HAC standard deviation
sharpeScreening(rets, control = list(nCore = 1, type = 2, hac = TRUE))

Testing the difference of Sharpe ratios

Description

Function which performs the testing of the difference of Sharpe ratios.

Usage

sharpeTesting(x, y, control = list())
sharpeTesting(x, y, control = list())

Arguments

x

Vector (of length $T$ ) of returns for the first fund. NA values are allowed.

y

Vector (of length $T$ ) returns for the second fund. NA values are allowed.

control

Control parameters (see *Details*).

Details

The Sharpe ratio (Sharpe 1994) is one industry standard for measuring the absolute risk adjusted performance of hedge funds. This function performs the testing of Sharpe ratio difference for two funds using the approach by Ledoit and Wolf (2002).

For the testing, only the intersection of non-NA observations for the two funds are used.

The argument control is a list that can supply any of the following components:

'type' Asymptotic approach (type = 1) or studentized circular bootstrap approach (type = 2). Default: type = 1.
'ttype' Test based on ratio (type = 1) or product (type = 2). Default: type = 2.
'hac' Heteroscedastic-autocorrelation consistent standard errors. Default: hac = FALSE.
'nBoot' Number of bootstrap replications for computing the p-value. Default: nBoot = 499.
'bBoot' Block length in the circular bootstrap. Default: bBoot = 1, i.e. iid bootstrap. bBoot = 0 uses optimal block-length.
'pBoot' Symmetric p-value (pBoot = 1) or asymmetric p-value (pBoot = 2). Default: pBoot = 1.

Value

A list with the following components:

n: Number of non-NA concordant observations.

sharpe: Vector (of length 2) of unconditional Sharpe ratios.

dsharpe: Sharpe ratios difference.

tstat: t-stat of Sharpe ratios differences.

pval: pvalues of test of Sharpe ratios differences.

Note

Further details on the methodology with an application to the hedge fund industry is given in Ardia and Boudt (2018).

Some internal functions where adapted from Michael Wolf MATLAB code.

Author(s)

David Ardia and Kris Boudt.

References

Ardia, D., Boudt, K. (2015). Testing equality of modified Sharpe ratios. Finance Research Letters 13, 97–104.

Ardia, D., Boudt, K. (2018). The peer performance ratios of hedge funds. Journal of Banking and Finance 87, 351–368.

Barras, L., Scaillet, O., Wermers, R. (2010). False discoveries in mutual fund performance: Measuring luck in estimated alphas. Journal of Finance 65(1), 179–216.

Sharpe, W.F. (1994). The Sharpe ratio. Journal of Portfolio Management 21(1), 49–58.

Ledoit, O., Wolf, M. (2008). Robust performance hypothesis testing with the Sharpe ratio. Journal of Empirical Finance 15(5), 850–859.

Storey, J. (2002). A direct approach to false discovery rates. Journal of the Royal Statistical Society B 64(3), 479–498.

Examples

## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
x = hfdata[,1]
y = hfdata[,2]

## Run Sharpe testing (asymptotic)
ctr = list(type = 1)
out = sharpeTesting(x, y, control = ctr)
print(out)

## Run Sharpe testing (asymptotic hac)
ctr = list(type = 1, hac = TRUE)
out = sharpeTesting(x, y, control = ctr)
print(out)

## Run Sharpe testing (iid bootstrap)
set.seed(1234)
ctr = list(type = 2, nBoot = 100)
out = sharpeTesting(x, y, control = ctr)
print(out)

## Run Sharpe testing (circular bootstrap)
set.seed(1234)
ctr = list(type = 2, nBoot = 100, bBoot = 5)
out = sharpeTesting(x, y, control = ctr)
print(out)
## Load the data (randomized data of monthly hedge fund returns)
data("hfdata")
x = hfdata[,1]
y = hfdata[,2]

## Run Sharpe testing (asymptotic)
ctr = list(type = 1)
out = sharpeTesting(x, y, control = ctr)
print(out)

## Run Sharpe testing (asymptotic hac)
ctr = list(type = 1, hac = TRUE)
out = sharpeTesting(x, y, control = ctr)
print(out)

## Run Sharpe testing (iid bootstrap)
set.seed(1234)
ctr = list(type = 2, nBoot = 100)
out = sharpeTesting(x, y, control = ctr)
print(out)

## Run Sharpe testing (circular bootstrap)
set.seed(1234)
ctr = list(type = 2, nBoot = 100, bBoot = 5)
out = sharpeTesting(x, y, control = ctr)
print(out)

Summary method for the 'SCREENING' object

Description

Produces a ranked summary of a screening: distribution of the performance measure, the peer performance ratios, optional win/loss counts from the pairwise tests, and the top/bottom funds (overall and by out-/underperformance ratio). Originally contributed by Murilo Andre Peres Pereira.

Usage

## S3 method for class 'SCREENING'
summary(object, coef = 1L, top = 5L, p_level = 0.05, ...)
## S3 method for class 'SCREENING'
summary(object, coef = 1L, top = 5L, p_level = 0.05, ...)

Arguments

object

A SCREENING object.

coef

For a screen_beta screening, which coefficient row to use (default 1, the alpha).

top

Number of best/worst funds to display. Default: 5.

p_level

Significance level for the win/loss counts (used only when the object carries square pval/tstat matrices). Default: 0.05.

...

Further arguments (ignored).

Value

An object of class summary.SCREENING.

Author(s)

Murilo Andre Peres Pereira, David Ardia and Kris Boudt.

Examples


  data("hfdata")
  sc <- alphaScreening(hfdata[, 1:10], control = list(nCore = 1))
  summary(sc)

data("hfdata")
  sc <- alphaScreening(hfdata[, 1:10], control = list(nCore = 1))
  summary(sc)

Targeted peer-performance screening for selected funds

Description

Performs peer-performance screening for a user-selected subset of funds (the “focal” funds) against the whole universe, returning detailed outputs for those focal funds only. It is a convenience wrapper that calls the corresponding screening function with the cross-group argument Y set to the full matrix X; equivalently, targetPeerPerformance(X, funds, "alpha") is alphaScreening(X[, funds], Y = X).

Usage

targetPeerPerformance(
  X,
  funds,
  method = c("alpha", "sharpe", "msharpe"),
  factors = NULL,
  level = 0.9,
  na.neg = TRUE,
  control = list()
)
targetPeerPerformance(
  X,
  funds,
  method = c("alpha", "sharpe", "msharpe"),
  factors = NULL,
  level = 0.9,
  na.neg = TRUE,
  control = list()
)

Arguments

X

Matrix $(T \times N)$ of $T$ returns for $N$ funds. NA values are allowed.

funds

Integer indices or character column names identifying the focal funds. The output preserves the order provided in funds.

method

Screening method: "alpha" (default), "sharpe", or "msharpe".

factors

Optional matrix of factor returns (only when method = "alpha").

level

Modified Value-at-Risk level (only when method = "msharpe").

na.neg

Logical; return NA for a negative modified Value-at-Risk (only when method = "msharpe").

control

Control parameters (see alphaScreening).

Value

A SCREENING object (see alphaScreening) whose fund-level components (pizero, pipos, pineg, the performance measure, npeer, ...) cover the focal funds only, and whose pval/tstat/difference matrices have one row per focal fund and one column per universe fund. Works with the print, plot, summary, and as.data.frame methods.

Author(s)

Murilo Andre Peres Pereira, David Ardia and Kris Boudt.

Examples

data("hfdata")
rets <- hfdata[, 1:10]

## focal Sharpe screening (funds by index) against the whole universe
out <- targetPeerPerformance(rets, funds = c(2, 5, 7), method = "sharpe",
                             control = list(nCore = 1))
out$sharpe
data("hfdata")
rets <- hfdata[, 1:10]

## focal Sharpe screening (funds by index) against the whole universe
out <- targetPeerPerformance(rets, funds = c(2, 5, 7), method = "sharpe",
                             control = list(nCore = 1))
out$sharpe

Package 'PeerPerformance'

Help Index

Screening using the alpha outperformance ratio

Description

Usage

Arguments

Details

Value

Note

Author(s)

References

See Also

Examples

Testing the difference of alpha outperformance ratios

Description

Usage

Arguments

Details

Value

Note

Author(s)

References

See Also

Examples

Coerce a 'SCREENING' object to a data frame

Description

Usage

Arguments

Value

Author(s)

See Also

Examples

Bootstrap confidence intervals for the peer performance ratios

Description

Usage

Arguments

Value

Author(s)

See Also

Examples

Factor exposure heterogeneity from a beta screening

Description

Usage

Arguments

Value

Author(s)

References

See Also

Examples

Hedge fund data

Description

Usage

Format

Source

Compute modified Sharpe ratio

Description

Usage

Arguments

Details

Value

Author(s)

References

See Also

Examples

Screening using the modified Sharpe outperformance ratio

Description

Usage

Arguments

Details

Value

Note

Author(s)

References

See Also

Examples

Testing the difference of modified Sharpe ratios

Description

Usage

Arguments

Details