Answers are grounded in this paper and cite their sources.
Let there be two samples X₁, X₂, ⋯, Xₘ and Y₁, Y₂, ⋯, Yₙ(N = m + n) from two populations with continuous cdf's $F(x)$ and $G(y)$. Let the first i ordered observations (out of N combined observations) contain mᵢx's and nᵢy's (mᵢ + nᵢ = i) where mᵢ and nᵢ are random numbers. To test {equation*}{0.1}H_0:F = G{equation*} against alternative that they are different we propose the statistic {equation*}{0.2}S(N)_r = ∑^ri = 1 a_iz_i + (m - m_r)(N - r)⁻¹(∑^Nᵣ₊₁a_i) - 1/2(m + n){equation*} based on the first r ordered observations only where aNᵢ = aᵢ = ∑Nj=N-i+1 1/j, and {align}z_i &= 1, if the ith ordered observation is an x_i, \\ &= 0, otherwise.{align} The statistic is the asymptotically most powerful rank test for censored data under the Lehmann alternative and is equivalent to the Savage statistic [14] when $r = N$. It is also known to maximize the minimum power over IFRA (or IFR) distributions asymptotically. Exact and large sample properties of S(N)ᵣ are studied and a k-sample extension of it is also considered. Various tables are also provided to facilitate the use of the S(N)ᵣ statistic.
No takes yet. Share an insight, caveat, or question.
Asit P. Basu (1968) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: