Confidence Intervals for Treatment Effect from Restricted Maximum Likelihood
Andrew L. Rukhin
Confidence intervals for the treatment effect in random effects meta-analysis model obtained from Harville-Jeske-Kenward-Roger approach are obtained in explicit form. They are compared to some other intervals commonly used in collaborative studies with a small number of participants possibly with heterogeneous, study-specific variances. Monte Carlo simulation experiments recommend the latter intervals.
Confidence Intervals for Treatment Effect from Restricted Maximum Likelihood, Computational Statistics, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=908146
(Accessed December 3, 2023)