Journal of Statistical Distributions and Applications

Table 5 For the data of Example 2.1, the adaptive scheme chose the score function with α =−12

From: Efficient and adaptive rank-based fits for linear models with skew-normal errors

α	β ₁	β ₂	β ₃	${\hat{τ}}_{φ}$	RobustR₂
α=−12 (Adaptive Choice)	0.0035	0.2190	0.0368	0.3675	0.2743
α=−10	0.0030	0.2095	0.0480	0.3699	0.2732
α=−9	0.0037	0.1923	0.0514	0.3531	0.2828
α=−8	0.0042	0.1831	0.0544	0.3484	0.2865
α=−7	0.0043	0.1822	0.0537	0.3533	0.2849
α=−6	0.0044	0.1808	0.0540	0.3490	0.2884
α=−5	0.0044	0.1805	0.0531	0.3580	0.2838
α=−4	0.0043	0.1854	0.0449	0.3886	0.2668

This table shows the variation in the rank-based fits for score functions in a neighborhood of α=−8.

Back to article page