Exact (version 1.5)

power.exact.test: Power calculation for unconditional exact test

Description

Calculates the power of the design for known sample sizes and true probabilities.

Usage

power.exact.test(p1, p2, n1, n2, npNumbers = 100, alpha = 0.05, 
                 alternative = "two.sided", interval = FALSE, beta = 0.001, 
                 method = "Z-pooled", ref.pvalue = TRUE, simulation = FALSE, 
                 nsim = 100)

Arguments

p1

The probability of success given in group 1

p2

The probability of success given in group 2

n1

The sample size in group 1

n2

The sample size in group 2

npNumbers

Number: The number of nuisance parameters considered

alpha

Significance level

alternative

Indicates the alternative hypothesis: must be either "less", "two.sided", or "greater"

interval

Logical: Indicates if a confidence interval on the nuisance parameter should be computed

beta

Number: Confidence level for constructing the interval of nuisance parameters considered. Only used if interval=TRUE

method

Indicates the method for finding tables as or more extreme than the observed table: must be either "Z-pooled", "Z-unpooled", "Santner and Snell", "Boschloo", "CSM", "CSM modified", or "CSM approximate"

ref.pvalue

Logical: Indicates if p-value should be refined by maximizing the p-value function after the nuisance parameter is selected

simulation

Logical: Indicates if the power calculation is exact or estimated by simulation

nsim

Number of simulations run. Only used if simulation=TRUE

Value

• The function returns the computed power.

Details

The power calculations are for binomial models. The design must know the fixed sample sizes in advance. There are $(n_1+1) \times (n_2+1)$ possible tables that could be produced. There are two ways to calculate the power: simulate the tables under two independent binomial distributions or consider all possible tables and calculate the exact power. The calculations can be done for any exact.test computation or using Fisher's exact test.

References

Berger, R. (1994) Power comparison of exact unconditional tests for comparing two binomial proportions. Institute of Statistics Mimeo Series No. 2266 Berger, R. (1996) More powerful tests from confidence interval p values. American Statistician, 50, 314-318 Boschloo, R. D. (1970), Raised Conditional Level of Significance for the 2x2-table when Testing the Equality of Two Probabilities. Statistica Neerlandica, 24, 1-35

See Also

statmod

Examples

power.exact.test(0.20,0.80,10,20)
power.exact.test(0.20,0.80,10,20,method="Fisher")
set.seed(218461)
power.exact.test(0.20,0.80,10,20,interval=TRUE,method="Boschloo",simulation=TRUE,nsim=100)