To produce p-values, the fit could be run once with rho fixed to zero. The posterior samples can then be used to generate many data sets under the same conditions (number of data points, error bar sizes). These can be refitted with rho free.
Comparing a summary statistic (such as the mean) on the |rho| posterior will give a p-value.
To produce p-values, the fit could be run once with rho fixed to zero. The posterior samples can then be used to generate many data sets under the same conditions (number of data points, error bar sizes). These can be refitted with rho free.
Comparing a summary statistic (such as the mean) on the |rho| posterior will give a p-value.