pst_sim: Power analysis for projected score test
In carolynlou/pstestr: Performs power analysis for the projected score test

Conducts power analysis for PST along with several other methods for range of user-specified mbetas, ks, and ns. User uses this function to conduct simulation study. Offers option for parallelization. Wraps and combines results from sim_setup() and pstest()

pst_sim(nsim = 500, seed = 2019, mbetas = c(0, 0.005, 0.01, 0.015,
  0.02, 0.025, 0.03, 0.035, 0.04, 0.045, 0.05, 0.06, 0.07, 0.08, 0.09,
  0.1), ks = c(40, 60, 70, 80, 100), n = 100, p = 1000,
  model = "normal", sigma = 1, alpha = 0.05, rho = 0.9,
  betasp = TRUE, rs = c(10, 20, 50), mc.cores = 1, doplot = TRUE)

`nsim`	number of simulations to conduct to assess power, defaults to 500
`seed`	chosen seed for simulations, defaults to 2019
`mbetas`	vector of mean beta values
`ks`	vector of percentage of independent variables with nonzero signal
`n`	number of observations, defaults to 100
`p`	number of betas, defaults to 1000
`model`	can be specified as 'normal' (default) for linear regression, otherwise does logistic regression
`sigma`	defaults to 1
`alpha`	significance level, defaults to 0.05
`rho`	spatial correlation in G parameter, AR1 structure, efaults to 0.9
`betasp`	indicator of presence of spatial information, defaults to TRUE
`rs`	investigator-specified set of "contrasts" of G, defaults to c(10, 20, 50)
`mc.cores`	number of cores to run on, defaults to 1
`doplot`	if TRUE, makes plots; if FALSE, does not. This produces one power plot per k across a range of mbetas