On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other
Citations
9,365 citations
Cites background or methods from "On a Test of Whether one of Two Ran..."
...Festinger 171, Haldane and Smith [14], Mann and Whitney [ 28 ], White [59], and Wilcoxon [61, 621 have published tables for the two-sample case....
[...]
...a An extensive comparison of exaot probabilities for the twc-aample test [ 28 ] with those based on the normal approximation indicates that the normal approximation is usually better with the continuity sdjustment when the probability is above 0.02, and better without it when the probability is 0.02 or below....
[...]
9,057 citations
Cites methods from "On a Test of Whether one of Two Ran..."
...The LEfSe algorithm is introduced in overview in the Results section, and Figure 6 illustrates in detail the format of the input (a matrix with n rows and m columns) and the three steps performed by the computational tool: the KW rank sum test [49] on classes, the pairwise Wilcoxon test [50,51] between subclasses of different classes, and the LDA [52] on the relevant features....
[...]
...for example, given some known population structure within a set of input samples, is a feature more abundant in all population subclasses or in just one? Specifically, we first use the non-parametric factorial KruskalWallis (KW) sum-rank test [49] to detect features with significant differential abundance with respect to the class of interest; biological consistency is subsequently investigated using a set of pairwise tests among subclasses using the (unpaired) Wilcoxon rank-sum test [50,51]....
[...]
8,655 citations
7,536 citations
Cites background from "On a Test of Whether one of Two Ran..."
...potential missing data, the psychometric properties of the variables examined, and the magnitude of the relationships considered before deciding on an appropriate sample size to use or to ensure that a sufficient sample size is actually available to study the phenomena of interest’’ (Marcoulides & Saunders, 2006, p. VI). Marcoulides and Saunders offer other warnings that clearly echo those presented in the CBSEM literature. For instance, SchneeweiX (2001) addresses the magnitude of standard errors in PLS path modeling estimators resulting from not using enough observations (consistency) and indicators for each latent variable (consistency at large)....
[...]
5,282 citations
References
12,871 citations
"On a Test of Whether one of Two Ran..." refers methods in this paper
...+ - (2n m + 2nm2 -n2 _ m2 -nm) 16 which is obtained from (1) by multiplication by (U nm/2)4 and expansion....
[...]
...Using the recurrence relation (1) the probabilities pnm(U) have been tabulated for m < n < 8 (see Table I)....
[...]
...After multiplying (1) by (U nm/2)2, using...
[...]
827 citations
"On a Test of Whether one of Two Ran..." refers background in this paper
...Since the probability of the ith 1 preceding the jth 0 is 4, we have (2) En,m(U) = nm/2....
[...]