calcNumRejects: Calculate number of cases rejected for repeated simulation of...
In PsumtSim: Simulations of grouped responses relative to baseline.

Description Usage Arguments Value Author(s) See Also Examples

Calculate number of cases rejected for repeated simulation of Poisson background and responses grouped into categories.

1	calcNumRejects(bkg, resps, numRespsPerCat, numSims, calcPValFnc, sigLevel = 0.05, ...)

`bkg`	Mean firing rate during a background interval, unrelated to stimulus presentation.
`resps`	Mean firing rates during a response period, one for each category.
`numRespsPerCat`	Number of repetitions (presentation of stimuli) in each category.
`numSims`	Number of simulation to run.
`calcPValFnc`	Function to be called with simulation output from `simCatResp` to calculate the achieved significance level, or p-value.
`sigLevel`	Significance level to use when determining if test is significant. Default is 0.05.
`...`	Other arguments to pass to `calcPValFnc`

Number of simulations which were detected as significant, out of numSims

Peter N. Steinmetz <PeterNSteinmetz@steinmetz.org>

compPowerRespDetection, compPowerGeneralRespDetection, simCatResp

  # Calculate number of cases which will be detected as having an effect of 
  # category when there are 4 categories with 2 having different responses and
  # when use a standard F test to detect the category effect.
  pvalFnc<-function(df){anova(glm(resp~category,data=df),test='F')$"Pr(>F)"[2]}
  calcNumRejects(1,c(1,1.5,2,1),6,100,pvalFnc)