kdeFSBT function

Full Significance Bayesian Testing

Performs Full Significance Bayesian Testing (FSBT) for univariate sharp null hypothesis based on a posterior sample. The marginal posterior density is obtained by kernel density estimation from sim.sample.

kdeFSBT(H0, sim.sample)

Arguments

• H0: a scalar value under the null hypothesis.
• sim.sample: a sample from the marginal posterior distribution.

Returns

double

Examples

x <-  rnorm(1000, 0, 1)
kdeFSBT(-1, x)

References

Pereira, C. A. d. B., Stern, J. M. and Wechsler, S. (2008) Can a significance test be genuinely Bayesian? Bayesian Analysis 3 , 79-100.