Table 2.

The MMMS and RSD (in parenthesis) of the simulated examples for linear and logistic regression model II using different number of PCs with n = 500 when p = 1000 and p = 10000.

PCs	Variance	SIS-PCA-MLR	SIS-PCA-MMLE	SIS-PCA-MLR	SIS-PCA-MMLE
Setting 1, linear model with p = 1000
		s = 6, β★ = (0.3,−0.3, 0.3, …)T		s = 12, β★ = (3, 4, 3, 4, …)T
0	0	12(35)	12(35)	30(21)	30(21)
1	18.4%	14(37)	84(66)	30.5(21)	31.5(23)
3	45.3%	19.5(32)	177.5(88)	30(19)	34.5(28)
5	63.6%	7(2)	53(26)	12(0)	39(16)
10	64.8%	7(3)	54(28)	12(1)	38(18)
30	68.9%	9(10)	63.5(40)	12(1)	35(15)
50	72.5%	14(15)	72(51)	12(1)	30.5(15)
100	79.8%	52.5(75)	119(109)	13(2)	28(15)
Setting 2, logistic regression with p = 1000
		s = 6, β★ = (0.7,−0.7, 0.7, …)T		s = 12, β★ = (−3, 4,−3, 4, …)T
0	0	13(37)	13(37)	856.5(241)	856.5(241)
1	18.7%	14.5(39)	81(73)	862(216)	879(193)
3	46.8%	19(37)	171(90)	878.5(165)	922.5(111)
5	64.7%	7(3)	50(26)	14(5)	73(35)
10	65.8%	7(3)	51(30)	15(7)	76(40)
30	69.9%	9(8)	59(39)	19(16)	84(49)
50	73.4%	12(15)	66(49)	25.5(33)	97(63)
100	80.6%	58(96)	132.5(127)	67(79)	139(98)
Setting 3, linear model with p = 10000
		s = 6, β★ = (0.3,−0.3, 0.3, …)T		s = 12, β★ = (3, 4, 3, 4, …)T
0	0	126.5(381)	126.5(381)	165.5(190)	165.5(190)
1	18.4%	154(357)	831(705)	168(186)	190(226)
3	46.8%	190.5(316)	1832.5(734)	154.5(186)	255(330)
5	64.6%	16(22)	489(242)	12(1)	179(113)
10	65.2%	16(29)	490(289)	12(1)	201(109)
30	67.3%	21.5(48)	549(325)	12(1)	206(118)
50	69.3%	34.5(74)	646(392)	12(2)	238(122)
100	74.0%	105(184)	860(590)	14(6)	284.5(162)
Setting 4, logistic regression with p = 10000
		s = 6, β★ = (0.7,−0.7, 0.7, …)T		s = 12, β★ = (−3, 4,−3, 4, …)T
0	0	163.5(302)	163.5(302)	7978.52840	7978.52840
1	2.0%	160(307)	240.5(327)	8122(2625)	8147.5(2583)
3	5.3%	167.5(300)	312(336)	8039.5(2382)	8092(2325)
5	7.4%	15(26)	51(29)	39(63)	77(40)
10	8.8%	15.5(27)	52(31)	41.5(57)	78.5(39)
30	14.1%	21(45)	56.5(36)	50(91)	85.5(46)
50	19.3%	30(65)	65.5(42)	71(118)	95(61)
100	31.3%	57(125)	87(65)	139(244)	128.5(127)