Calculates the power curve (d x power) for an experiment with a fixed
number of instances. See calc_instances() for details.
calc_power_curve(ninstances, sig.level = 0.05, alternative = "two.sided", test.type = "t.test", d.range = c(0.1, 2), npoints = 100)
Arguments
| ninstances | the number of instances to be used in the experiment. |
|---|---|
| sig.level | significance level (alpha) for the experiment |
| alternative | type of alternative hypothesis ("two.sided" or "one.sided") |
| test.type | type of test ("t.test", "wilcoxon", "binomial") |
| d.range | vector |
| npoints | number of points for the power curve. |
Value
an object of class caiser.powercurve containing fields
d, the (standardized) effect size; and power, the (expected) power
to detect each effect size in d.
Examples
my.cpc <- calc_power_curve(ninstances = 10) summary(my.cpc)#> #==================================== #> Number of points: 100 #> power = 0.5 for d = 0.69 #> power = 0.8 for d = 1 #> power = 0.95 for d = 1.3 #> #====================================plot(my.cpc)#> #==================================== #> Number of points: 100 #> power = 0.5 for d = 0.69 #> power = 0.8 for d = 1 #> power = 0.95 for d = 1.3 #> #====================================
