# Check LEP1 TPC's ridge yield upper limit
## [Slide (p.3-p.10): before bin-by-bin correction](https://www.dropbox.com/s/pojkhbdfn5dvkp1/20220308.pdf?dl=0)
1. p.3-p.4: The reproduced raw correlations agree with the archived paper results in general.
3. p.6: The upper limits differ much from the paper, especially for the beam axis results.
4. p.7-p.10: For more information, the ZYAM-fits and bootstrap distributions are shown.
> In the bootstrap plots, the legend should be ALEPH, √s=91.2 GeV.
> From top to bottom, the results are fourier fits, cos+polynomial and polynomial.

- Some possibilties that could cause the difference:
- [ ] ZYAM-fit templates are slightly different from those of LEP1 paper. We removed odd terms under the assumption that correlation functions are symmetric about $\Delta \phi=0$.
- [ ] Bin-by-bin correction haven't been applied
> See below
- [x] ZYAM fit range
> $\Delta \phi: (0,\pi/2)$ is the nominal fit range
- [ ] a difference in the pseudodata generation

## [Slide: after bin-by-bin correction + using the codes on the master branch](https://www.dropbox.com/s/qwcw1jw702rs18b/CheckLEP1UL.pdf?dl=0)
1. Using `/afs/cern.ch/user/a/abaty/public/forAnthony/April5_2019_crosschecks/` for the bin-by-bin correction ([`compareTPCP.cc`](https://github.com/yenjie/StudyMult/blob/master/TwoParticleCorrelation/compareTPCP.cc#L13-L25)) on the archived paper correlation results.
2. Using the codes `doFit.C`, `bootstrapConfInterval.C` and `Systematics2PC.h` on the master branch:
> Grendel paths:
> - `/home/janicechen/StudyMult/TwoParticleCorrelation/src/doFit.C`
> - `/home/janicechen/StudyMult/TwoParticleCorrelation/include/bootstrapConfInterval.C`
> - `/home/janicechen/StudyMult/TwoParticleCorrelation/include/Systematics2PC.h`
4. p.2: After correction, still see a large difference from the paper fig.3.
5. p.3,4: The bootstrap distributions are shown.
> But I think they look qualitatively similar as the ones on Anthony's thesis?