-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathresult.txt
More file actions
115 lines (102 loc) · 10.4 KB
/
Copy pathresult.txt
File metadata and controls
115 lines (102 loc) · 10.4 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
=== ZERO-NOISE: flat_calm ===
[IV solver] n=136 mean_err=0.011034 max_err=0.033494 rejected=0
[SVI fit] true: a=0.020000 b=0.080000 rho=-0.100000 m=0.000000 sigma=0.200000
expiry=2024-01-16 fit: a=0.019952 b=0.080160 rho=-0.094539 m=0.000643 sigma=0.200199 |da|=0.000048 |db|=0.000160 |drho|=0.005461 rmse_vol_pts=0.067286 converged=1 arb_free=1
expiry=2024-02-01 fit: a=0.019955 b=0.080150 rho=-0.094644 m=0.000630 sigma=0.200184 |da|=0.000045 |db|=0.000150 |drho|=0.005356 rmse_vol_pts=0.046020 converged=1 arb_free=1
expiry=2024-03-02 fit: a=0.019961 b=0.080132 rho=-0.094834 m=0.000608 sigma=0.200156 |da|=0.000039 |db|=0.000132 |drho|=0.005166 rmse_vol_pts=0.031878 converged=1 arb_free=1
expiry=2024-04-01 fit: a=0.019966 b=0.080115 rho=-0.095019 m=0.000586 sigma=0.200129 |da|=0.000034 |db|=0.000115 |drho|=0.004981 rmse_vol_pts=0.025372 converged=1 arb_free=1
[signals] count=0 (expect ~0 at zero noise, scaling up with injected noise)
=== REALISTIC-NOISE: flat_calm ===
[IV solver] n=136 mean_err=0.010983 max_err=0.042099 rejected=0
[SVI fit] true: a=0.020000 b=0.080000 rho=-0.100000 m=0.000000 sigma=0.200000
expiry=2024-01-16 fit: a=0.015313 b=0.088252 rho=-0.092694 m=-0.001507 sigma=0.235227 |da|=0.004687 |db|=0.008252 |drho|=0.007306 rmse_vol_pts=0.350571 converged=1 arb_free=1
expiry=2024-02-01 fit: a=0.025104 b=0.068556 rho=-0.132272 m=-0.005305 sigma=0.159578 |da|=0.005104 |db|=0.011444 |drho|=0.032272 rmse_vol_pts=0.234000 converged=1 arb_free=1
expiry=2024-03-02 fit: a=0.007036 b=0.104697 rho=-0.044112 m=0.009862 sigma=0.276438 |da|=0.012964 |db|=0.024697 |drho|=0.055888 rmse_vol_pts=0.143726 converged=1 arb_free=1
expiry=2024-04-01 fit: a=0.011078 b=0.099912 rho=0.030728 m=0.030897 sigma=0.248727 |da|=0.008922 |db|=0.019912 |drho|=0.130728 rmse_vol_pts=0.112120 converged=1 arb_free=1
[signals] count=0 (expect ~0 at zero noise, scaling up with injected noise)
=== ZERO-NOISE: steep_skew ===
[IV solver] n=136 mean_err=0.012911 max_err=0.041271 rejected=0
[SVI fit] true: a=0.030000 b=0.120000 rho=-0.350000 m=0.000000 sigma=0.150000
expiry=2024-01-16 fit: a=0.029961 b=0.120158 rho=-0.347060 m=0.000172 sigma=0.150128 |da|=0.000039 |db|=0.000158 |drho|=0.002940 rmse_vol_pts=0.063397 converged=1 arb_free=1
expiry=2024-02-01 fit: a=0.029962 b=0.120153 rho=-0.347122 m=0.000168 sigma=0.150123 |da|=0.000038 |db|=0.000153 |drho|=0.002878 rmse_vol_pts=0.043347 converged=1 arb_free=1
expiry=2024-03-02 fit: a=0.029964 b=0.120142 rho=-0.347235 m=0.000161 sigma=0.150116 |da|=0.000036 |db|=0.000142 |drho|=0.002765 rmse_vol_pts=0.030005 converged=1 arb_free=1
expiry=2024-04-01 fit: a=0.029966 b=0.120132 rho=-0.347345 m=0.000154 sigma=0.150108 |da|=0.000034 |db|=0.000132 |drho|=0.002655 rmse_vol_pts=0.023861 converged=1 arb_free=1
[signals] count=0 (expect ~0 at zero noise, scaling up with injected noise)
=== REALISTIC-NOISE: steep_skew ===
[IV solver] n=136 mean_err=0.012844 max_err=0.048723 rejected=0
[SVI fit] true: a=0.030000 b=0.120000 rho=-0.350000 m=0.000000 sigma=0.150000
expiry=2024-01-16 fit: a=0.025987 b=0.128858 rho=-0.325500 m=-0.000576 sigma=0.171789 |da|=0.004013 |db|=0.008858 |drho|=0.024500 rmse_vol_pts=0.337711 converged=1 arb_free=1
expiry=2024-02-01 fit: a=0.033124 b=0.111988 rho=-0.381845 m=-0.002520 sigma=0.133833 |da|=0.003124 |db|=0.008012 |drho|=0.031845 rmse_vol_pts=0.225661 converged=1 arb_free=1
expiry=2024-03-02 fit: a=0.024602 b=0.133600 rho=-0.296598 m=0.003995 sigma=0.174337 |da|=0.005398 |db|=0.013600 |drho|=0.053402 rmse_vol_pts=0.142205 converged=1 arb_free=1
expiry=2024-04-01 fit: a=0.024610 b=0.134284 rho=-0.257664 m=0.011608 sigma=0.171127 |da|=0.005390 |db|=0.014284 |drho|=0.092336 rmse_vol_pts=0.111322 converged=1 arb_free=1
[signals] count=0 (expect ~0 at zero noise, scaling up with injected noise)
=== ZERO-NOISE: high_curvature ===
[IV solver] n=136 mean_err=0.011556 max_err=0.041797 rejected=0
[SVI fit] true: a=0.015000 b=0.200000 rho=-0.250000 m=0.000000 sigma=0.080000
expiry=2024-01-16 fit: a=0.014984 b=0.200126 rho=-0.248579 m=0.000042 sigma=0.080039 |da|=0.000016 |db|=0.000126 |drho|=0.001421 rmse_vol_pts=0.064719 converged=1 arb_free=1
expiry=2024-02-01 fit: a=0.014985 b=0.200122 rho=-0.248606 m=0.000041 sigma=0.080038 |da|=0.000015 |db|=0.000122 |drho|=0.001394 rmse_vol_pts=0.044232 converged=1 arb_free=1
expiry=2024-03-02 fit: a=0.014985 b=0.200115 rho=-0.248657 m=0.000039 sigma=0.080037 |da|=0.000015 |db|=0.000115 |drho|=0.001343 rmse_vol_pts=0.030593 converged=1 arb_free=1
expiry=2024-04-01 fit: a=0.014986 b=0.200108 rho=-0.248706 m=0.000037 sigma=0.080035 |da|=0.000014 |db|=0.000108 |drho|=0.001294 rmse_vol_pts=0.024312 converged=1 arb_free=1
[signals] count=0 (expect ~0 at zero noise, scaling up with injected noise)
=== REALISTIC-NOISE: high_curvature ===
[IV solver] n=136 mean_err=0.011504 max_err=0.049210 rejected=0
[SVI fit] true: a=0.015000 b=0.200000 rho=-0.250000 m=0.000000 sigma=0.080000
expiry=2024-01-16 fit: a=0.014368 b=0.201610 rho=-0.248111 m=-0.000296 sigma=0.083000 |da|=0.000632 |db|=0.001610 |drho|=0.001889 rmse_vol_pts=0.329400 converged=1 arb_free=1
expiry=2024-02-01 fit: a=0.015778 b=0.197369 rho=-0.251105 m=-0.000171 sigma=0.077120 |da|=0.000778 |db|=0.002631 |drho|=0.001105 rmse_vol_pts=0.222895 converged=1 arb_free=1
expiry=2024-03-02 fit: a=0.014005 b=0.204060 rho=-0.241122 m=0.000631 sigma=0.083366 |da|=0.000995 |db|=0.004060 |drho|=0.008878 rmse_vol_pts=0.139191 converged=1 arb_free=1
expiry=2024-04-01 fit: a=0.013634 b=0.205874 rho=-0.227583 m=0.002122 sigma=0.084090 |da|=0.001366 |db|=0.005874 |drho|=0.022417 rmse_vol_pts=0.110002 converged=1 arb_free=1
[signals] count=0 (expect ~0 at zero noise, scaling up with injected noise)
=== ZERO-NOISE: low_vol_regime ===
[IV solver] n=136 mean_err=0.006865 max_err=0.025012 rejected=0
[SVI fit] true: a=0.005000 b=0.060000 rho=-0.150000 m=0.000000 sigma=0.120000
expiry=2024-01-16 fit: a=0.004955 b=0.060182 rho=-0.139376 m=0.001004 sigma=0.120260 |da|=0.000045 |db|=0.000182 |drho|=0.010624 rmse_vol_pts=0.096435 converged=1 arb_free=1
expiry=2024-02-01 fit: a=0.004960 b=0.060160 rho=-0.139633 m=0.000982 sigma=0.120222 |da|=0.000040 |db|=0.000160 |drho|=0.010367 rmse_vol_pts=0.065832 converged=1 arb_free=1
expiry=2024-03-02 fit: a=0.004969 b=0.060120 rho=-0.140092 m=0.000942 sigma=0.120153 |da|=0.000031 |db|=0.000120 |drho|=0.009908 rmse_vol_pts=0.045493 converged=1 arb_free=1
expiry=2024-04-01 fit: a=0.004977 b=0.060083 rho=-0.140523 m=0.000905 sigma=0.120089 |da|=0.000023 |db|=0.000083 |drho|=0.009477 rmse_vol_pts=0.036179 converged=1 arb_free=1
[signals] count=0 (expect ~0 at zero noise, scaling up with injected noise)
=== REALISTIC-NOISE: low_vol_regime ===
[IV solver] n=136 mean_err=0.007132 max_err=0.040035 rejected=0
[SVI fit] true: a=0.005000 b=0.060000 rho=-0.150000 m=0.000000 sigma=0.120000
expiry=2024-01-16 fit: a=0.004873 b=0.059742 rho=-0.148013 m=-0.000457 sigma=0.123188 |da|=0.000127 |db|=0.000258 |drho|=0.001987 rmse_vol_pts=0.487092 converged=1 arb_free=1
expiry=2024-02-01 fit: a=0.006226 b=0.056446 rho=-0.147915 m=0.000070 sigma=0.105545 |da|=0.001226 |db|=0.003554 |drho|=0.002085 rmse_vol_pts=0.333379 converged=1 arb_free=1
expiry=2024-03-02 fit: a=0.002760 b=0.066932 rho=-0.127189 m=0.000981 sigma=0.141345 |da|=0.002240 |db|=0.006932 |drho|=0.022811 rmse_vol_pts=0.191142 converged=1 arb_free=1
expiry=2024-04-01 fit: a=0.003032 b=0.066859 rho=-0.072396 m=0.010847 sigma=0.136179 |da|=0.001968 |db|=0.006859 |drho|=0.077604 rmse_vol_pts=0.152553 converged=1 arb_free=1
[signals] count=0 (expect ~0 at zero noise, scaling up with injected noise)
=== STAGE 4: degenerate case handling ===
[deep ITM call, 1y] true_sigma=0.200000 recovered=0.268001 converged=1 iters=21
[deep OTM call, 1y] true_sigma=0.200000 recovered=0.000101 converged=1 iters=1
[ATM, 1 day] true_sigma=0.200000 recovered=0.200000 converged=1 iters=2
[deep OTM put, 1 wk] true_sigma=0.200000 recovered=0.000101 converged=1 iters=1
=== STAGE 5: injected mispricing detection sweep ===
bg_noise offset(vp) injected detected false_pos total_pts
zero 0.500000 12 0 0 136
zero 1.000000 12 0 0 136
zero 2.000000 12 0 0 136
zero 3.000000 12 12 0 136
zero 5.000000 12 12 0 136
zero 8.000000 12 12 0 136
realistic 0.500000 12 0 0 136
realistic 1.000000 12 0 0 136
realistic 2.000000 12 0 0 136
realistic 3.000000 12 12 0 136
realistic 5.000000 12 12 0 136
realistic 8.000000 12 12 0 136
[note] detected/injected gives true-positive rate at each
offset; false_pos should stay ~0. The offset where detected
first reaches injected consistently is your practical
detection threshold under that noise regime.
=== SUMMARY (8 scenario runs) ===
scenario n_pts rej mean_iv_err max_iv_errmax_param_err signals converged
ZERO-NOISE: flat_calm 136 0 0.011034 0.033494 0.005461 0 yes
REALISTIC-NOISE: flat_calm 136 0 0.010983 0.042099 0.130728 0 yes
ZERO-NOISE: steep_skew 136 0 0.012911 0.041271 0.002940 0 yes
REALISTIC-NOISE: steep_skew 136 0 0.012844 0.048723 0.092336 0 yes
ZERO-NOISE: high_curvature 136 0 0.011556 0.041797 0.001421 0 yes
REALISTIC-NOISE: high_curvature 136 0 0.011504 0.049210 0.022417 0 yes
ZERO-NOISE: low_vol_regime 136 0 0.006865 0.025012 0.010624 0 yes
REALISTIC-NOISE: low_vol_regime 136 0 0.007132 0.040035 0.077604 0 yes
worst max_iv_err across all scenarios: 0.049210
worst max_param_err across all scenarios: 0.130728
any non-converged fit: no
[done] This table is your validation evidence. Screenshot it
and cite the worst-case numbers, not the best-case ones.