GRS.test-package {GRS.test}

GRS Test for Portfolio Efficiency, Its Statistical Power Analysis, and Optimal Significance Level Calculation

Description

Computational resources for test proposed by Gibbons, Ross, Shanken (1989)<DOI:10.2307/1913625>. It also has the functions for the power analysis and the choice of the optimal level of significance. The optimal level is determined by minimizing the expected loss from hypothesis testing.

Details

The DESCRIPTION file:

Package:	GRS.test
Type:	Package
Title:	GRS Test for Portfolio Efficiency, Its Statistical Power Analysis, and Optimal Significance Level Calculation
Version:	1.2
Date:	2022-06-29
Author:	Jae H. Kim <[e-mail]>
Maintainer:	Jae H. Kim <[e-mail]>
Description:	Computational resources for test proposed by Gibbons, Ross, Shanken (1989)<DOI:10.2307/1913625>. It also has the functions for the power analysis and the choice of the optimal level of significance. The optimal level is determined by minimizing the expected loss from hypothesis testing.
License:	GPL-2

Index of help topics:

GRS.MLtest              GRS Test Statistic and p-value based on Maximum
                        Likelihood Estimator for Covariance matrix
GRS.Power               Statistical Power of the GRS test
GRS.Powerfunc           Power functions for the GRS test
GRS.T                   Sample Size Selection for the GRS test
GRS.optimal             Optimal Level of Significance for the GRS test:
                        Normality Assumption
GRS.optimalboot         Optimal Level of Significance for the GRS test:
                        Bootstrapping
GRS.optimalbootweight   Weighted Optimal Level of Significance for the
                        GRS test: Bootstrapping
GRS.optimalweight       Weighted Optimal Level of Significance for the
                        GRS test: Normality Assumption
GRS.test                GRS test and Model Estimation Results
GRS.test-package        GRS Test for Portfolio Efficiency, Its
                        Statistical Power Analysis, and Optimal
                        Significance Level Calculation
data                    Fama-French Data: 25 size-B/M portfolio and
                        risk factors, obtained from French's library

The package accompanies the working paper:

Kim and Shamsuddin, 2017, Empirical Validity of Asset-pricing Models: Application of Optimal Significance Level and Equal Probability Test

The function GRS.test returns the GRS test statistics with model estimation results.

The function GRS.MLtest provides an alternative test statistic with theta and theta* estimation results.

Additional functions for the power analysis and calculation of optimal level of significance are also included.

Author(s)

Jae H. Kim <[e-mail]>

Maintainer: Jae H. Kim <[e-mail]>

References

Gibbons, Ross, Shanken, 1989. A test of the efficiency of a given portfolio, Econometrica, 57,1121-1152. <DOI:10.2307/1913625>

Fama and French, 1993, Common risk factors in the returns on stocks and bonds, Journal of Financial Economics, 33, 3-56. <DOI:10.1016/0304-405X(93)90023-5>

Fama and French, 2015, A five-factor asset-pricing model, Journal of Financial Economics, 1-22. <DOI:http://dx.doi.org/10.1016/j.jfineco.2014.10.010>

See Also

The examples replicate the results reported in Fama and French (1993) and Kim and Shamsuddin (2016)

Examples

data(data)
factor.mat = data[1:342,2:4]            # Fama-French 3-factor model
ret.mat = data[1:342,8:ncol(data)]      # 25 size-BM portfolio returns
GRS.test(ret.mat,factor.mat)$GRS.stat   # Table 9C of Fama-French (1993)