Two-Stage Adaptive Optimal Design with Fixed First-Stage Sample Size

Table 2

Integrated absolute difference of the cumulative distributions (×100) of 𝑇₁∼𝒩(0,𝑀⁻¹(𝜉^∗,𝜃)), 𝑇₂∼𝒩(0,𝑀⁻¹(𝜉_𝐴,𝜃)), and 𝑇₃∼𝑈𝑄 versus the approximate cumulative distribution of √𝑛(𝜃_𝑛−𝜃) obtained via Monte Carlo simulations for various 𝑛₁ and various large sizes of 𝑛. The values 𝜃=1, 𝑥₁=2, 𝜎=.5, 𝑎=.25, and 𝑏=4 were used.

𝑛1	𝑛=400		𝑛=1000
	𝑇2	𝑇3	𝑇2	𝑇3
50	13	9	13	5
100	10	9	8	4
200	11	14	4	7