testH0: Generate a Bayesian p-value for a Hypothesis Test
In tmelliott/bsm: Bayesian Selectivity Modelling of Fisheries Trawl Data

Description Usage Arguments Details Value Author(s)

Computes the posterior probability of the null hypothesis being true.

1	testH0(fit, parameter, H0, alternative = "!=")

`fit`	a bsmfit object
`parameter`	the name of the parameter to test
`H0`	the null hypothesis value
`alternative`	the alternative hypotheis, one of "!=", "<" or ">"

The parameter has a posterior distribution, and this function computes the posterior probability that the parameter is consistently larger or smaller than the hypothesised value, which ever is smaller,

min(Pr(theta > H0), Pr(theta < H0))

In the case of a one-sided test, the probability is simply the posterior probability that the parameter is greater than, or smaller than, the hypothesised value.

numeric

Tom Elliott

tmelliott/bsm documentation built on May 31, 2019, 4:38 p.m.