>And so, the researchers came up with the burden of proof risk function, a novel statistical method to quantitatively “evaluate and summarize evidence of risk across different risk-outcome pairs.”
My stats knowledge is admittedly shallow, but this seemed a bit out of place to me. Is it common to develop novel statistical vehicles for meta-studies like this? My expectation would be that this is a kind of investigation we've been doing for a while, so new models aren't really needed.
My stats knowledge is admittedly shallow, but this seemed a bit out of place to me. Is it common to develop novel statistical vehicles for meta-studies like this? My expectation would be that this is a kind of investigation we've been doing for a while, so new models aren't really needed.