Heavy-tailed data make covariance estimation sensitive to the clipping level: stronger clipping reduces variance but increases bias. We study how to choose this clipping level from the data within a radial clipped covariance family. We propose the quantile tail-energy surrogate (QTES), a fully data-driven rule that combines a cross-fitted variance certificate with a held-out estimate of the tail energy removed by clipping. QTES requires no distributional prior parameters. For Euclidean clipping, the operator-norm bias is bounded by this scalar tail-energy quantity. Under a finite L4 moment condition, together with a mild feasibility condition for the QTES block construction, we prove a uniform finite-sample calibration bound over a positive clipping grid and a finite-sample guarantee for the estimator selected by QTES, relative to the best candidate on that grid under the same variance-plus-bias criterion. Experiments on clean heavy-tailed and contaminated benchmarks show that QTES performs strongly across both regimes and remains competitive with the best reference methods considered.
Figures & tables
Distribution
QTES
Random- γ
Oracle
Sel./Oracle
Top-2 Hit
Gaussian
0.3330
0.3384
0.3122
1.0660
0.5000
Student- t(5)
0.3335
0.3410
0.3130
1.0657
0.5000
Lognormal
0.3492
0.3713
0.3296
1.0585
0.5800
Table 1: Selector diagnostics at (n,d)=(400,200) over 100 trials per distribution
Gaussian
Student- t(5)
Lognormal
Method
RelSpecErr
SubspaceErr
Time (s)
RelSpecErr
SubspaceErr
Time (s)
RelSpecErr
SubspaceErr
Time (s)
QTES
0.3330
0.2625
0.0300
0.3335
0.2618
0.0303
0.3492
0.2768
0.0269
SCM
0.3534
0.2621
0.0004
0.3618
0.2640
0.0004
0.5994
0.3665
0.0004
OAS
0.3919
0.2621
0.0081
0.3909
0.2640
0.0081
0.5045
0.3665
0.0075
MCD
0.4101
0.3021
0.6514
0.4138
0.3045
0.6474
0.4139
0.3421
0.7239
WM-oracle
0.3487
0.2622
0.0007
0.3451
0.2621
0.0007
0.3417
0.2797
0.0007
Table 2: Clean-data benchmark at (n,d)=(400,200)
Figure 1: Sample-size scaling of the relative spectral error of QTES and the WM-oracle at fixed dimension d=200 . The vertical dashed line indicates the phase transition threshold n=d . Shaded regions represent ±1 standard deviation over 100 independent trials. Both axes are on a logarithmic scale.
Method
Trials
c=1
c=25
c=100
c=400
QTES
100
0.3484
0.3833
0.3911
0.3873
SCM
100
0.5867
3.5147
13.5848
56.1519
OAS
100
0.4824
3.1133
12.7698
53.5103
MCD
100
0.4116
0.4256
0.4270
0.4194
WM-oracle
100
0.3406
0.3563
0.3549
0.3547
WM-adaptive
100
0.4462
0.4424
0.4455
0.4460
Table 3: Gross-contamination stress test with standardized lognormal inliers at (n,d)=(400,200) . Entries are relative spectral errors.
Appendix figures & tables1 asset
Supplementary material from the paper’s appendix.
Appendix
Experimental Scenario
Mean Estimator
RelSpecErr
SubspaceErr
Gaussian (clean)
Sample mean
0.34991±0.03777
0.26875±0.01378
MOM estimator
0.35049±0.03770
0.26888±0.01378
Known mean reference
0.33300±0.03177
0.26251±0.01073
Student- t(5) (clean)
Sample mean
0.34912±0.03770
0.26882±0.01254
MOM estimator
0.34889±0.03785
0.26877±0.01246
Known mean reference
0.33349±0.03061
0.26183±0.01159
Appendix
Table 4: QTES with true mean μ=1d unknown to the adaptive procedures, over 100 independent trials at (n,d)=(400,200) . Entries report mean ± standard deviation.