Closed DHekstra closed 3 days ago
Without thinking too hard, I feel negative values should be a warning. I can't say for sure there is no legitimate use case for negative correlation coefficients. Greater than one being an error makes sense to me.
resolved by #170
motivated by some intrepid users: double-wilson parameters < 0 or >1 are not intended use cases and may lead to unexpected behavior. Perhaps the default should be to resort to univariate priors for negative r and throw an error for r>1?