power_diff | R Documentation |
There are two ways of operation:
power=NULL
Given sample size n
it simulates the power.
n=NULL
Given a power an iterative search is started to find a suitable n
within a specified range.
power_diff( distribution = c("exponential", "weibull"), param = "delay", test = c("bootstrap", "pearson", "moran", "lr", "lr_pp"), eff = stop("Provide parameters for both group that reflect the effect!"), n = NULL, r = 1, sig.level = 0.05, power = NULL, nPowerSim = 1600, R = 201, nRange = c(5, 50) )
distribution |
character. Which assumed distribution is used for the power calculation. |
param |
character. Parameter name(s) for which to simulate the power. |
test |
character. Which test to use for this power estimation? |
eff |
list. The two list elements contain the model parameters (as understood by the delay-distribution functions provided by this package) for the two groups. |
n |
integer. Number of observations per group for the power simulation or |
r |
numeric. Ratio of both groups sizes, ny / nx. Default value is 1, i.e., balanced group sizes. Must be positive. |
sig.level |
numeric. Significance level. Default is 0.05. |
power |
numeric. |
nPowerSim |
integer. Number of simulation rounds. Default value 1600 yields a standard error of 0.01 for power if the true power is 80%. |
R |
integer. Number of bootstrap samples for test of difference in parameter within each power simulation. It affects the resolution of the P-value for each simulation round. A value of around |
nRange |
integer. Admissible range for sample size when power is pre-specified and sample size is requested. |
In any case, the distribution, the parameters that are tested for, the type of test and the effect size (eff=
) need to be specified.
The more power simulation rounds (parameter nPowerSim=
) the more densely the space of data according to the specified model is sampled.
Note that this second modus (when n
is estimated) is computationally quite heavy.
The iterative search for n
uses some heuristics and the estimated sample size might actually give a different power-level.
It is important to check the stated power in the output. The search algorithm comes to results closer to the power aimed at
when the admissible range for sample size (nRange=
) is chosen sensibly.
In case the estimated sample size and the achieved power is too high it might pay off to rerun the function with an adapted admissible range.
List of results of power simulation. Or NULL
in case of errors.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.