The approximate behaviour and performance of the two-sample trimmed t

By K. Yuen, W. Dixon. 1973

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Few-Shot Out-of-Domain Transfer Learning of Natural Language Explanations (2021, methodology): “For every model comparison, we report the statistical significance via the paired Student’s t-test for equal variances [Yuen and Dixon, 1973], with single-tailed p-values and 0.”
Confidence Interval Coverage for Cohen's Effect Size Statistic (2006, methodology): “To do so, replace the usual t statistic by the statistic of Yuen and Dixon (1973), t h h h h Y Y S R t t= + −⎛ ⎝⎜ ⎞ ⎠⎟ 1 2 1 2 2 1 ~ , where hj = nj – 2gj and ~ ( ) ,S N S h h W2 2 1 2 2 2 = − + − where hj is the number of observations remaining after trimming.”