GRS.test (version 1.0)

GRS.optimal: Optimal Level of Significance for the GRS test

Description

The optimal level is calculated by minimizing expected loss from hypothesis testing

Under the assumption of equal prior and identical losses from Type I and II errors

Usage

GRS.optimal(T, N, K, theta, ratio, Graph = "TRUE")

Arguments

T

sample size

N

the number of portfolio returns

K

the number of risk factors

theta

maximum Sharpe ratio of the K factor portfolios

ratio

theta/thetas, proportion of the potential efficiency

Graph

show graph if TRUE. No graph otherwise

Value

Details

Based on the power calculation of the GRS test, as in GRS (1989) .

The blue square is the point where the expected loss is mimimized.

The red horizontal line indicate the point of the covnentional level of significance (alpha = 0.05).

References

Leamer, E. 1978, Specification Searches: Ad Hoc Inference with Nonexperimental Data, Wiley, New York.

Kim, JH and Ji, P. 2015, Significance Testing in Empirical Finance: A Critical Review and Assessment, Journal of Empirical Finance 34, 1-14.

Gibbons, Ross, Shanken, 1989. A test of the efficiency of a given portfolio, Econometrica, 57,1121-1152.

Kim and Shamsuddin, 2016, Reapparaising Empirical Validity of Asset-Pricing Models with consideration of Statistical Power. Working Paper

See Also

Kim and Ji (2015)

Examples

GRS.optimal(T=90, N=25, K=3, theta=0.25, ratio=0.4) # Figure 3 of Kim and Shamsuddin (2016)

Run the code above in your browser using DataCamp Workspace