From: A simulation study for comparing testing statistics in response-adaptive randomization
Optimal allocation ratio (n _{1}/n _{2}) for maximizing powers | |
---|---|
R_{ Risk } | |
R _{ Odds }/R_{ Chisq } | |
R _{ Wald }/R_{ Neyman } | |
R_{ LLR } | {q _{2} - p _{2} exp[I _{1} - I _{2}/(p _{2} - p _{1})]}/{-q _{1} + p _{1} exp[I _{1} - I _{2}/(p _{2} - p _{1})]} |
Other allocation targets | |
R _{ RPW }/R_{ DL } | q _{2}/q _{1} |
R_{ RSIHR } | (Minimize the number of failure at fixed power of T _{ Wald }) |