A New Method for Setting Futures Portfolios’ Maintenance Margins: Evidence from Chinese Commodity Futures Markets

Table 6

Effectiveness of our method under different relative risk aversion coefficients.

	Interval	𝑎	PI	OCI
Portfolio A	In-sample	0.5	0.9674	0.0627
		0.6	0.9832	0.0671
		0.7	1.0000	0.0778
		0.8	1.0000	0.0815
		0.9	1.0000	0.0860
	Out-of-sample	0.5	0.9476	0.0501
		0.6	0.9657	0.0615
		0.7	0.9983	0.0709
		0.8	0.9994	0.0810
		0.9	1.0000	0.0852
Portfolio B	In-sample	0.5	0.9356	0.0684
		0.6	0.9474	0.0575
		0.7	0.9792	0.06844
		0.8	1.0000	0.0858
		0.9	1.0000	0.0893
	Out-of-sample	0.5	0.9364	0.0674
		0.6	0.9582	0.0797
		0.7	0.9993	0.0847
		0.8	0.9995	0.0889
		0.9	1.0000	0.0904
Portfolio C	In-sample	0.5	0.9467	0.0649
		0.6	0.9770	0.0666
		0.7	0.9867	0.0748
		0.8	1.0000	0.0806
		0.9	1.0000	0.0835
	Out-of-sample	0.5	0.9476	0.0501
		0.6	0.9657	0.0615
		0.7	0.9845	0.0694
		0.8	0.9906	0.0795
		0.9	1.0000	0.0847
Portfolio D	In-sample	0.5	0.9575	0.0585
		0.6	0.9807	0.0697
		0.7	0.9897	0.0745
		0.8	1.0000	0.0873
		0.9	1.0000	0.0906
	Out-of-sample	0.5	0.9578	0.0567
		0.6	0.9667	0.0675
		0.7	0.9975	0.0693
		0.8	0.9981	0.0795
		0.9	1.0000	0.0814

Notes: 𝑎 stands for the relative risk aversion coefficient.