Abstract
In the two-sample location problem, if the t-statistic is computed as a function of the ranks of the observations rather than the observations themselves, the resulting statistic, TR, is a function of the Wilcoxon statistic. It is shown here that TR has approximately a t-distribution. The ARE of the TR -test is identical to that of the Wilcoxon test. A Monte Carlo study reveals that the TR-test is more robust, with regard to the significance level, than the t-test.
| Original language | English |
|---|---|
| Pages (from-to) | 383-394 |
| Number of pages | 12 |
| Journal | Communications in Statistics Part B: Simulation and Computation |
| Volume | 10 |
| Issue number | 4 |
| DOIs | |
| State | Published - Jan 1 1981 |
| Externally published | Yes |
Fingerprint
All Science Journal Classification (ASJC) codes
- Modeling and Simulation
- Statistics and Probability
Cite this
The rank transform in the two-sample location problem. / Nath, Ravinder; Duran, B. S.
In: Communications in Statistics Part B: Simulation and Computation, Vol. 10, No. 4, 01.01.1981, p. 383-394.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - The rank transform in the two-sample location problem
AU - Nath, Ravinder
AU - Duran, B. S.
PY - 1981/1/1
Y1 - 1981/1/1
N2 - In the two-sample location problem, if the t-statistic is computed as a function of the ranks of the observations rather than the observations themselves, the resulting statistic, TR, is a function of the Wilcoxon statistic. It is shown here that TR has approximately a t-distribution. The ARE of the TR -test is identical to that of the Wilcoxon test. A Monte Carlo study reveals that the TR-test is more robust, with regard to the significance level, than the t-test.
AB - In the two-sample location problem, if the t-statistic is computed as a function of the ranks of the observations rather than the observations themselves, the resulting statistic, TR, is a function of the Wilcoxon statistic. It is shown here that TR has approximately a t-distribution. The ARE of the TR -test is identical to that of the Wilcoxon test. A Monte Carlo study reveals that the TR-test is more robust, with regard to the significance level, than the t-test.
UR - http://www.scopus.com/inward/record.url?scp=0039874375&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0039874375&partnerID=8YFLogxK
U2 - 10.1080/03610918108812216
DO - 10.1080/03610918108812216
M3 - Article
VL - 10
SP - 383
EP - 394
JO - Communications in Statistics Part B: Simulation and Computation
JF - Communications in Statistics Part B: Simulation and Computation
SN - 0361-0918
IS - 4
ER -