sims-uasettings: Simulation study results (unimodal, Gaussian-distributed)

In stephaniehicks/benchmarkfdrData2019: Data and Benchmarking Results from Korthauer and Kimes et al. (2019)

Description

Results for simulations used in Korthauer and Kimes et al. (2019) with 'unimodal' series of effect size distributions (Stephens, 2017) with standard Gaussian distributed test statistics, and ‘cubic’ informative and paired uninformative covariates stored as a list of SummarizedBenchmark objects from 100 replications. In each simulation, 10% of tests were non-null, i.e. had non-zero effect sizes. Simulations were performed with ‘bimodal’, ‘flat top’, ‘skew’ and ‘spiky’ effect size distributions (see corresponding reference for details).

In sims-uasettings-t and sims-uasettings-25, similar settings were considered, but with t_{11} distributed test statistics and 25% of tests being non-null, respectively.

Arguments

metadata

logical whether to load metadata (TRUE) or full resource (FALSE). (default = FALSE)

Details

The object is a list of length 100 corresponding to the 100 replications. Each entry in the list is a named list of two SummarizedBenchmark objects. The results based on using an informative and uninformative covariate are stored as "informative" and "uninformative", respectively.

References

• Stephens, M. (2017). False discovery rates: a new deal. Biostatistics, 18(2), 275-294.

See Also

sims-uasettings-t, sims-uasettings-25

Examples

`uasettings-benchmark-bimodal`(metadata = TRUE)

## Not run: 
`uasettings-benchmark-bimodal`()
`uasettings-benchmark-flattop`()
`uasettings-benchmark-skew`()
`uasettings-benchmark-spiky`()

res <- `uasettings-benchmark-spiky`()
res[[1]]$informative
res[[1]]$uninformative

## End(Not run)

stephaniehicks/benchmarkfdrData2019 documentation built on June 20, 2021, 10 a.m.