Skip to content

Conversation

@cschwan
Copy link
Contributor

@cschwan cschwan commented May 1, 2021

Result for the TTM_TRAP grid:

   PineAPPL         MC        sigma      central         min      max 
                              1/100   sigma   1/1000   1/1000   1/1000
----------------------------------------------------------------------
 1.311377e+01  1.311299e+01   0.053   0.114   0.0597   0.0577   0.0642
 1.664682e+01  1.664020e+01   0.043   0.920   0.3978   0.4278   0.3561
 1.412514e+01  1.412249e+01   0.050   0.373   0.1879   0.1955   0.1743
 8.545668e+00  8.545810e+00   0.093   0.018   0.0166   0.0354   0.0105
 2.168749e+01  2.167976e+01   0.045   0.801   0.3565   0.3880   0.3246
 2.651202e+01  2.650650e+01   0.059   0.356   0.2083   0.2325   0.1841
 2.200552e+01  2.199958e+01   0.118   0.229   0.2701   0.2909   0.2489
 1.347035e+01  1.347064e+01   0.214   0.010   0.0215   0.0336   0.0067
 1.269538e+01  1.269187e+01   0.081   0.342   0.2767   0.3019   0.2474
 1.547207e+01  1.546577e+01   0.071   0.578   0.4076   0.4446   0.3598
 1.346821e+01  1.346327e+01   0.080   0.462   0.3670   0.3945   0.3304
 9.406538e+00  9.406787e+00   0.073   0.036   0.0265   0.0408   0.0047
 5.107627e+00  5.106697e+00   0.164   0.111   0.1820   0.1940   0.1752
 6.478143e+00  6.475772e+00   0.131   0.280   0.3661   0.4010   0.3223
 7.092014e+00  7.087261e+00   0.154   0.436   0.6706   0.7257   0.5956
 6.476083e+00  6.475440e+00   0.163   0.061   0.0994   0.1031   0.0967

@cschwan cschwan requested a review from enocera May 1, 2021 13:46
@cschwan
Copy link
Contributor Author

cschwan commented May 1, 2021

Result for TTM_TTRAP grid:

   PineAPPL         MC        sigma      central         min      max 
                              1/100   sigma   1/1000   1/1000   1/1000
----------------------------------------------------------------------
 1.347101e+01  1.347292e+01   0.040   0.354   0.1413   0.1685   0.1007
 1.384937e+01  1.383888e+01   0.048   1.580   0.7576   0.8260   0.6615
 1.105290e+01  1.105132e+01   0.054   0.267   0.1437   0.1427   0.1404
 1.415802e+01  1.415825e+01   0.054   0.030   0.0164   0.0276   0.0001
 2.304689e+01  2.304497e+01   0.035   0.239   0.0834   0.0975   0.0753
 2.336406e+01  2.334589e+01   0.038   2.032   0.7783   0.8593   0.6777
 1.797411e+01  1.797534e+01   0.026   0.269   0.0686   0.0959   0.0293
 1.929493e+01  1.929490e+01   0.040   0.004   0.0015   0.0029   0.0217
 1.548244e+01  1.547073e+01   0.056   1.346   0.7568   0.8405   0.6487
 1.534003e+01  1.533694e+01   0.049   0.409   0.2017   0.2149   0.1837
 1.106589e+01  1.106609e+01   0.058   0.030   0.0174   0.0310   0.0010
 9.156840e+00  9.156946e+00   0.081   0.014   0.0116   0.0312   0.0213
 8.848567e+00  8.839714e+00   0.095   1.051   1.0015   1.1035   0.8624
 8.343414e+00  8.343454e+00   0.084   0.006   0.0048   0.0076   0.0015
 5.229780e+00  5.229836e+00   0.098   0.011   0.0107   0.0080   0.0098
 2.730001e+00  2.730055e+00   0.188   0.011   0.0198   0.0862   0.0848

@cschwan
Copy link
Contributor Author

cschwan commented May 3, 2021

Result for the TRAP_TPT grid:

   PineAPPL         MC        sigma      central         min      max 
                              1/100   sigma   1/1000   1/1000   1/1000
----------------------------------------------------------------------
 1.869506e+01  1.869158e+01   0.063   0.297   0.1860   0.1946   0.1775
 1.969838e+01  1.969385e+01   0.045   0.509   0.2302   0.2540   0.2099
 1.041951e+01  1.041600e+01   0.089   0.379   0.3369   0.3702   0.2964
 2.833407e+00  2.831908e+00   0.145   0.365   0.5290   0.5808   0.4585
 2.330940e+01  2.330415e+01   0.053   0.424   0.2253   0.2418   0.2045
 2.513230e+01  2.512417e+01   0.047   0.693   0.3238   0.3592   0.2857
 1.316896e+01  1.316286e+01   0.096   0.480   0.4632   0.5057   0.4053
 3.477437e+00  3.476103e+00   0.122   0.315   0.3839   0.4113   0.3444
 2.107228e+01  2.106358e+01   0.048   0.853   0.4129   0.4432   0.3677
 2.205031e+01  2.204278e+01   0.053   0.647   0.3414   0.3668   0.3131
 1.094877e+01  1.094684e+01   0.076   0.231   0.1757   0.1784   0.1697
 2.608017e+00  2.607770e+00   0.118   0.080   0.0945   0.0917   0.0963
 1.578711e+01  1.578712e+01   0.067   0.001   0.0009   0.0161   0.0184
 1.493146e+01  1.493160e+01   0.084   0.012   0.0098   0.0157   0.0109
 6.235262e+00  6.235161e+00   0.093   0.017   0.0161   0.0004   0.0394
 1.044806e+00  1.044769e+00   0.224   0.016   0.0350   0.0108   0.0695

@cschwan
Copy link
Contributor Author

cschwan commented May 3, 2021

Plot for TRAP_TPT (0 < yt < 0.35):

output-0

TRAP_TPT (0.35 < yt < 0.85):

output-1

TRAP_TPT (0.85 < yt < 1.45):

output-2

TRAP_TPT (1.45 < yt < 2.5):

output-3

@cschwan
Copy link
Contributor Author

cschwan commented May 3, 2021

Plot for TTM_TRAP (340 < Mtt < 400):

output-0

TTM_TRAP (400 < Mtt < 500):

output-1

TTM_TRAP (500 < Mtt < 650):

output-2

TTM_TRAP (650 < Mtt < 1500):

output-3

@cschwan
Copy link
Contributor Author

cschwan commented May 3, 2021

Plot for TTM_TTRAP (340 < Mtt < 400):

output-0

TTM_TTRAP (400 < Mtt < 500):

output-1

TTM_TTRAP (500 < Mtt < 650):

output-2

TTM_TTRAP (650 < Mtt < 1500):

output-3

@enocera enocera force-pushed the CMS_TTB_8TEV_2D branch from d1ad377 to 03797e0 Compare May 3, 2021 17:54
Copy link
Contributor

@enocera enocera left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

May I suggest to change the y-axis label in the first inset of all of the plots in order to clarify that the distribution is double differential? Example (for the first plot):
dsigma/dyt -> dsigma/dytdpTt

@cschwan
Copy link
Contributor Author

cschwan commented May 31, 2021

@enocera The previous commit makes all distributions two dimensional.

Results:

bin    yt       ptt   d2sig/dyt/dptt neg unc pos unc
---+----+----+---+---+--------------+-------+-------
  0    0 0.35   0  80   6.6768066e-1 -12.98%  12.99%
  1    0 0.35  80 150   8.0401565e-1 -12.69%  12.17%
  2    0 0.35 150 250   2.9770022e-1 -13.17%  12.77%
  3    0 0.35 250 600   2.3129849e-2 -13.98%  14.02%
  4 0.35 0.85   0  80   5.8273500e-1 -12.85%  12.81%
  5 0.35 0.85  80 150   7.1806573e-1 -12.69%  12.17%
  6 0.35 0.85 150 250   2.6337918e-1 -13.21%  12.81%
  7 0.35 0.85 250 600   1.9871069e-2 -14.02%  14.04%
  8 0.85 1.45   0  80   4.3900574e-1 -12.91%  12.85%
  9 0.85 1.45  80 150   5.2500729e-1 -12.86%  12.42%
 10 0.85 1.45 150 250   1.8247943e-1 -13.38%  13.02%
 11 0.85 1.45 250 600   1.2419128e-2 -14.26%  14.28%
 12 1.45  2.5   0  80   1.8794175e-1 -13.39%  13.36%
 13 1.45  2.5  80 150   2.0314908e-1 -13.55%  13.25%
 14 1.45  2.5 150 250   5.9383442e-2 -14.29%  14.18%
 15 1.45  2.5 250 600   2.8430084e-3 -15.12%  15.23%

output-0
output-1
output-2
output-3

bin   Mtt       yt     d2sig/dMtt/dyt neg unc pos unc
---+---+----+----+----+--------------+-------+-------
  0 340  400    0 0.35   6.2446513e-1 -12.97%  13.62%
  1 340  400 0.35 0.85   5.5489399e-1 -12.83%  13.39%
  2 340  400 0.85 1.45   3.9236515e-1 -12.63%  13.03%
  3 340  400 1.45  2.5   1.3564553e-1 -12.35%  12.21%
  4 400  500    0 0.35   6.1964261e-1 -12.76%  12.49%
  5 400  500 0.35 0.85   5.3024048e-1 -12.68%  12.38%
  6 400  500 0.85 1.45   3.6675874e-1 -12.59%  12.17%
  7 400  500 1.45  2.5   1.2828909e-1 -12.88%  12.43%
  8 500  650    0 0.35   2.4181679e-1 -13.20%  12.47%
  9 500  650 0.35 0.85   2.0629427e-1 -13.13%  12.38%
 10 500  650 0.85 1.45   1.4964678e-1 -13.34%  12.71%
 11 500  650 1.45  2.5   5.9724051e-2 -14.15%  14.02%
 12 650 1500    0 0.35   1.7168492e-2 -13.92%  12.70%
 13 650 1500 0.35 0.85   1.5242689e-2 -13.65%  12.29%
 14 650 1500 0.85 1.45   1.3905909e-2 -14.63%  14.10%
 15 650 1500 1.45  2.5   7.2561156e-3 -16.21%  16.75%

output-0
output-1
output-2
output-3

bin   Mtt       ytt    d2sig/dMtt/dytt neg unc pos unc
---+---+----+----+----+---------------+-------+-------
  0 340  400    0 0.35    6.4147674e-1 -13.03%  13.69%
  1 340  400 0.35 0.75    5.7705702e-1 -12.86%  13.45%
  2 340  400 0.75 1.15    4.6053768e-1 -12.64%  13.11%
  3 340  400 1.15  2.5    1.7479035e-1 -12.38%  12.36%
  4 400  500    0 0.35    6.5848258e-1 -12.83%  12.61%
  5 400  500 0.35 0.75    5.8410141e-1 -12.72%  12.47%
  6 400  500 0.75 1.15    4.4935276e-1 -12.63%  12.33%
  7 400  500 1.15  2.5    1.4292539e-1 -12.64%  11.98%
  8 500  650    0 0.35    2.9490355e-1 -13.43%  12.95%
  9 500  650 0.35 0.75    2.5566721e-1 -13.33%  12.79%
 10 500  650 0.75 1.15    1.8443155e-1 -13.30%  12.68%
 11 500  650 1.15  2.5    4.5218962e-2 -13.54%  12.66%
 12 650 1500    0 0.35    2.9743081e-2 -14.54%  14.01%
 13 650 1500 0.35 0.75    2.4539452e-2 -14.59%  14.01%
 14 650 1500 0.75 1.15    1.5381707e-2 -14.69%  14.04%
 15 650 1500 1.15  2.5    2.3790855e-3 -15.04%  14.17%

output-0
output-1
output-2
output-3

@cschwan cschwan requested a review from enocera May 31, 2021 11:11
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants