Subgroup Analysis of a Single-Index Threshold Penalty Quantile Regression Model Based on Variable Selection | Wuhan University Journal of Natural Sciences

Open Access

Table 1

Variable selection in the case where $ε_{i} \sim M V N (0, Σ)$ and $Σ$ follows an $A R$ -1 structure

Parameter	Quantile	Model	$n = 400$				$n = 800$
Parameter	Quantile	Model	TN	FN	Correct/%	MSE/10^-2	TN	FN	Correct/%	MSE/10^-2
$p = d = 10$	0.25	proposed	15	0	100	0.357	15	0	100	0.191
		oracle	15	0	100	0.357	15	0	100	0.191
	0.50	proposed	15	0	100	0.328	15	0	100	0.163
		oracle	15	0	100	0.332	15	0	100	0.160
	0.75	proposed	15	0	100	0.387	15	0	100	0.195
		oracle	15	0	100	0.366	15	0	100	0.193
$p = d = 20$	0.25	proposed	35	0	100	0.393	35	0	100	0.195
		oracle	35	0	100	0.357	35	0	100	0.191
	0.50	proposed	35	0	100	0.337	35	0	100	0.158
		oracle	35	0	100	0.332	35	0	100	0.160
	0.75	proposed	35	0	100	0.368	35	0	100	0.193
		oracle	35	0	100	0.366	35	0	100	0.193

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.