Calculates a prediction interval for the summary parameters of Riley's alternative model for bivariate random-effects meta-analysis. This interval predicts in what range future observations will fall given what has already been observed.
# S3 method for riley
predict(object, ...)Data frame containing prediction intervals with the summary estimates beta1 and beta2
(for effect size data), or with the mean sensitivity and false positive rate (for diagnostic test accuracy data).
A riley object.
Additional arguments (currently ignored)
Thomas Debray <thomas.debray@gmail.com>
Prediction intervals are based on Student's t-distribution with (numstudies - 5) degrees of freedom. The width of the interval is specified by the significance level chosen during meta-analysis.