In the bivariate meta-analysis of data on true positives and negative \(\mathrm{CT}\) scan diagnoses, adopt a scale mixture of normals for the \(\theta_{i}=\left(\theta_{i 1}, \theta_{i 2}\right)\) in the first model, namely

\[T P_{i} \sim \operatorname{Bin}\left(D_{i}, s_{i}\right),\]

with \(v\) an unknown, with prior \(\frac{1}{v} \sim U(0.5,0.02)\). Are any studies classed as outliers using this strategy?

Transcribed Image Text:

TN, ~Bin(W, 1), logit (s) = logit (t;) = 0,2; ~ BVN ; ~ Ga a().