power_2t_unequal {dvmisc}

Calculate Power for Two-Sample Unequal Variance T-Test

Description

Unequal variance version of power_2t_equal. Assumes an equal sample size for both groups, which is actually not optimal.

Usage

power_2t_unequal(n = 100, d, sigsq1, sigsq2, alpha = 0.05)

Arguments

n

Numeric value specifying per-group sample size.

d

Numeric value specifying true difference in group means. Should be positive.

sigsq1, sigsq2

Numeric value specifying the variance of observations in each group.

alpha

Numeric value specifying type-1 error rate.

Value

Numeric value.

Examples

# Power to detect difference of 0.2 with 100 subjects per group and sigsq's
# of 1 and 1.25
power_2t_unequal(n = 100, d = 0.2, sigsq1 = 1, sigsq2 = 1.25)