Comments from Mauro Donega

Type B comments (everything else, e.g. strategy, paper structure, emphasis, additions/subtractions, etc):

Line 160 / Line 271: The top reconstruction algorithm you use is recommended for pT > 500 while in the hadronic analysis it used down to a pT of 350 GeV. Have you checked the effects of this? Did you re-derive scale factors and uncertainties?

The top reconstruction algorithm is recommended for use pase 500 GeV, but validated down to 400 GeV. The choice of top pt was made before this top tagging decision. The increase from 350->400 would not have much impact on the analysis, but we think this is likely covered by the uncertainty so we would prefer to keep it as is .

Lines 236-239: why not fix the mass of the reconstructed top to the nominal top mass? it would probably increase the resolution on the signal mass. a kinematic fit would even be better

Since we make a selection on the reconstructed top mass we cannot fix to the nominal top mass. In the past alternative methods of mass reconstruction for the leptonic analysis have been investigated and found to not improve performance.

Line 240: This needs more details. Are the AK8 jets used with or without softdrop? The use of AK8 jets as b-jets is also problematic. At best this hurts the resolution. Do you have plots showing how the reconstruction works for tops? And the mass and pT of the bs reconstructed by the AK8 jets?

The AK8 4-vectors jets are not calculated using softdrop, only cuts on the mass for event selections are used. The reason for the AK8 b jets is essentially a loose anti top/W tag, ie if the cone is wide, then ttbar in this kinematic regime will be largely merged into an AK8 jet whereas signal will be an isolated b. This lets us largely eliminate most ttbar without needing the more complicated full anti-object tagging. Also, this lets us be confident that the energy scale is the same (ie two AK8 jets) when constructing the Mtb variable. I am confused about the other comment, this b is not used to reconstruct tops with (b+W), but rather is an isolated b from the W'. Attached are plots of the AK8 four vector jet mass and pt for the signal.

  • bjetmass.png:
    bjetmass.png

  • bpt.png:
    bpt.png

It is possible that the use of AK8 jets could slightly degrade the resolution, and we think that this would be a good analysis decision to review in the next iteration, ie. to see if the improved resolution can help the analysis sensitivity.

line 256: why does the variable pt(j1+j2) discriminate between signal and background? if the W' is produced at rest pt(t)=pt(b), and this variable is approximately pt(b)-1/3*pt(t) ~ 2/3*pt(b) or equivalently 2/3*pt(t). but you are already cutting on pt(t) so it is unclear how cutting on this other variable adds discrimination. please add an explanation or remove the cut

You are correct that this variable peaks around 2/3*pt(t) for signal, but for background this is not the case since the t and b candidates are not produced as back to back. This variable is a convenient way to encode both pt and angular information into one observable, and is found to improve sensitivity. We have added a clarification sentence to the text.

section 4.5: are the offline cuts in the plateaus of the chosen triggers? I'm particularly concerned about the HT > 800 GeV requirement. if not: how do you take into account the possible sculpting of the background shape? also: how were the trigger turn ons measured? what uncertainty is obtained on this measurement?

The measurement of the trigger efficiency is explained on L267. You are correct that the offline cuts do not put the analysis completely in the plateau, and so the measured turn on is applied as an event weight. The systematic uncertainty associated to this procedure us detailed in section 6.

line 282: why do you need to remove the bjet requirement to enrich the control sample in ttbar events?

It is not a matter of removing the requirement necessarily. We do not need the b-jet requirement in the dilepton region to enrich in ttbar, so we omit it in order to improve statistics.

section 5.1.1: the defined control region is kinematically very far from the signal region: how do you ensure that the results found here also apply to the signal region? is any extrapolation in pt done?

We use two different control regions, with different modifications from the signal selection. We find that the results agree in both control regions, and this gives us confidence that the results can be safely applied to the signal region. Note that no correction is derived in these regions, the effect of the existing top pt reweighting procedure is simply checked.

Line 317 and below: -How is it prevented that events are part of the signal and control region, but with switched roles? B-jets in the signal window have to be below 70 GeV, top jets in the control window between 50-170 GeV. So a signal-region B-jet could become a control region top jet if it has a mass between 50 and 70 GeV.

Right, but the tau32 inversion still ensures orthogonality %ENDCOLOR

-How is ttbar treated when estimating the Multijet background?

ttbar is subtracted when creating the b tagging rate and then the ttbar that passes the b tagging rate is subtracted from the QCD bkg estimate. In this way, the QCD estimate is purely from QCD origin. It should also be mentioned that these subtractions are very small, so the ttbar subtraction method is mainly implemented for consistency %ENDCOLOR

Table 2: The observed agreement between data and sum-of-backgrounds is extremely good and the uncertainty extremely small. It almost looks too good to be true. It would be helpful to improve and expand the current section 5.2.1 to better understand the procedures used.

We thought this looked strange as well and discovered that these numbers were out of date and missing the lepton veto etc (possibly from an svn merge issue). The table has been updated to the current analysis state. Also, to be consistent with the leptonic table, the hadronic channel has added the normalization uncertainties.

Type A comments (English/Style/Formatting, including figures):

abstract: “Two searches..”: Better “A search using two final states..”

Changed.

abstract: combination of the searches : combination of both searches

Changed.

abstract: “Results of the combination are also presented..” -> Repetition.. Maybe: Results are presented for the two channels individually as well as combined.

Changed.

abstract: define leptonic and hadronic analysis

This does not seem necessary to do in the abstract

abstract: lepton (e, mu): the letter e should not be italic

Changed.

abstract: "no evidence is found... and the production": remove and (repetition)

Done.

line 2: add (SM) after standard model, standard model should not be written in capitalized letters - see the CMS publication guidelines (https://twiki.cern.ch/twiki/bin/viewauth/CMS/Internal/PubGuidelines)

Done.

line 3: “these additional gauge bosons”: drop the gauge, clear from previous sentence. Otherwise make it “charged gauge”, otherwise confusing

Changed.

lines 3 and 4: "it is common... it is common": please rephrase (and avoid repetition)

Changed.

line 6: “desireable”: reword

Done.

line 6: “it is more distinguishable..when compared to..” - What is more distinguishable? Reword

Done.

line 8: W'->qq' remove italic from q

Done.

line 8: “also useful”: reword

Done.

line 9: W'->lv: the ell is not defined

Done.

line 9: "SM W or Z boson": remove the italic from the W and Z

Done.

Line 10: “also advantagous”: reword

Changed.

lines 10-11: "the tb decay channel... final state)" : rather confused sentence, please rephrase (a good start could be to replace 'determined' by something like 'measured', remove the 'a' in 'a leptonic final state, change 'leptonic' into semileptonic)

Changed

Line 16: “will present..” -> “presents..”

Done.

line 18: integrated luminosity of 2.6 fb-1: contradicts the abstract

Changed.

Line 18: the analyzed data correspond (remove the s), data is plural

Done.

line 19: the EXPECTED production cross section

Done.

line 20: resonances -> resonance

Changed.

line 21: we separately analyze -> we define two categories?

Analyze seems to make more sense here, categorization is potentially confusing with respect event categories and this is a much bigger difference. Left as-is.

line 22: "the invariant mass of the top and bottom quark": please rephrase and define better what is used (as is it seems that the top mass and the bottom mass are used)

Changed.

line 24: "the achieved sensitivity after the combination is higher than is possible in each individual channel": in this case, this states a just statistical truth, please remove.

This is still an important point to highlight (and the combination could still provide effectively no improvement).

line 25: provides -> and provides (?)

Changed.

section 2 should be renamed "The CMS detector and particle reconstruction" and the information now contained in sections 4.1 and 4.2 should be integrated here

We think the jet information is more suited to the object selection section and prefer to keep the current content organization.

line 31: awkward placement of ref [15]: what is the point of this reference? is it to define pseudorapidity? if so please add the sentence "The CMS detector and its reference system are described in detail elsewhere [15]" (or something like what said in lines 71-72) at the beginning of the section

Fixed.

line 35: "remaining barrel photons": remaining from what? please rephrase

This statement is taken from the official detector section twiki and is left as-is.

lines 44 and 45: "nonshowering" and "showering" are jargon, please rephrase

This statement is taken from the official detector section twiki and is left as-is.

lines 46-50 are awkward and should be removed

Done.

lines 56-67 are misplaced and contain several repetitions with respect to the text that preceeds them. please re-organize this whole section in such a way that the description of the particle flow reconstruction is not disjointed from the description of the objects

Done.

section 3.1: the information of which mass samples (with which granularity) have been generated is missing

Added.

Line 75: remove “then”

Done.

line 76: is the k-factor 1.2 really justified by the cited references? from ref 20 it seems that a factor 1.29 would be more suited

Reference 21 provides a more accurate determination as function of W' mass and here it can be seen that 1.2 is well suited to our mass range of interest.

Line 79: matrix -> element of the matrix

Changed.

line 80: g_W is labeled as g_w (lower case) in the formula. please harmonize

Using g_W.

lines 82-84: "tau" -> \tau

Done.

line 83: "no distincition is made in the analysis": this remark is confusing. what is the point here? please explain

This statement clarifies that we include electrons from W->tau nu->e nu nu decays in the analysis, not only W->e nu. This statement clarifies that we include electrons from W->tau nu->e nu nu decays in the analysis, not only W->e nu. Line 83: muon decaying directly?

"muon directly from the W boson decay"

line 88: Where does the right handed neutrino suddenly come from? Introduce properly.

Statement added.

Line 91: define much lighter

Changed.

line 95: "always assumed": not true (already the next sentence contradicts it)

Qualifiers before the statement are given to clarify this.

Line 100: analysis -> channel (twice)

"Analysis" left to remain consistent with the rest of the paper.

section 3.2: this section is very vague and should be removed, as all the relevant information is in section 5

This section gives important information about the simulation details, while section 5 covers the methods of estimation and correction, so the two should be kept separate.

section 3: this section should be renamed "Simulated samples" and only the description of the generators (signal and BG) should be given. no subsections are needed. the information of section 3.3 should be expanded, giving details on each subsample.

Subsection headers have been removed, but the content is left as-is.

line 100: please define the physical process before using the expressions "W+jets" and "tt"

This is clear already, should not be necessary.

Line 102, line 104: drop “as a background” and “as backgrounds”

Done.

Line 108: starting a sentence with a symbol is odd

Fixed.

the information contained in sections 4.1 and 4.2 should be moved to section 2

We have chosen to leave this information in section 4, and leave section 2 as exclusively the CMS Detector section.

Line 129 by ->with and specify the GEANT version

Changed.

Line 113: Explain more exactly the difference in phase space. Would assume the overlap with the phase space for signal to be similar?

A discussion of the phase space differences seems too detailed for the paper. The differences in reconstruction of the top quark and differences in top pT are most significant.

Line 133: No need to repeat again that there is a leptonic and hadronic analysis. Clear by now.

Removed.

Line 134: Remove “additional”

Done.

line 137 "both analyses use the charged hadron subtraction method" is jargon and is anyway an awkward way to begin a section on jet reconstruction

The method is described, and the method name is given simply for completeness. The section label is not present anymore since this text is now in the detector section.

line 138: "leading vertex" needs to be defined

Done.

line 139: "proton-proton pair interactions" remove "pair"

Done.

line 142: "final level": not clear final for what, please rephrase removing "final"

Statement removed.

line 145 and 146: please rephrase avoiding the word "prefer" ("uses" would already be better)

Done.

lines 150-157 seem superfluous and could be removed, or should be at least rephrased (for example removing the mention of the jet energy corrections, which was already done)

Restructured.

line 159 also here "both analyses" is an awkward way of beginning a section on identification of bjets

Rephrased.

line 159 what is a candidate jet? consider removing "candidate"

Removed.

line 159: it needs to be made more clear that the leptonic and hadronic analysis use different objects as input to b-tagging

Done.

line 160 "loose operating point" is jargon, rephrase

Done.

Line 160: an -> a

Done.

line 161 "correct observed differences" differences in what? what is corrected? please specify. also please quantify the magnitude of the differences

Clarified. The magnitude of the correction should not be necessary in the paper.

Line 161: Remove the hyphen between “b” and “tagging”. This is valid for the entire paper - this is not done consistently. Also b tagging is jargon, please explain before using (see CMS publication guidelines)

Done.

lines 164-165 "to account for small differences in tagging" -> "to account for small differences in tagging efficiency" (?) also please quantify the smallness of the differences

Actually this is just an additional uncertainty of 3% (scale factor consistent with 1). This has been clarified %ENDCOLOR

Line 166: Same comment than for b tagging

Done.

equation 2: if beta=0 why bother writing down that whole last parenthesis? please remove

Right, there is no R dependence. But this is the description of the general softdrop algorithm, which I think is still important for the reader along with the operating point that we use %ENDCOLOR

Line 171 (+2 lines): Do you really decluster AK8 jets? Softdrop procedure is first to recluster the constitutents with Cambridge/Aachen. Please verify and adapt text accordingly.

Right, good point. This has been changed %ENDCOLOR

line 173: if R0 is 0.8, why not write it?

Once again here we want to first state the general formula and then state the operating point. %ENDCOLOR

Line 174: beta=0 softdrop corresponds to modified mass-drop tagger. Mention and cite.

Done%ENDCOLOR

line after 175: "defines tau_N variables" -> "defines N variables"

Better to give variable symbol here.

line 176: again, if R=0.8 why not write it?, isn’t R the same than R0 above? If so, use the same symbol

As before, we prefer to keep the description of the general procedure as well as our specific operating point %ENDCOLOR

Line 176-184: too many uses of “consistent”. Reword.

Done.

Line 179: “all particle flow inputs”: Before or after CHS? Clarify

After, but we are not sure if this needs to be clarified as chs is mentioned earlier

Line 182: Rewrite the sentence saying that g/q jet are consistent with one, but you used the 3 to 2 ratio as discriminating variable

Done.

Line 184: “We select events..”. I assume you select JETS.

Fixed.

line 188 please add the top tagging efficiency for real tops

Done.

Line Cite: Quote: JME-15-002

Done.

line 189: "small differences" how small?

Added that the scale factors are consistent with 1

rename section 4.2 to "identification of electrons and muons"

Done.

lines 206-208 "both the leptonic... than in the leptonic analysis." : this choice is really bizarre and needs either a better justification, or the electron identification needs to be synchronized between the two channels

This choice is motivated by technical differences in the analysis frameworks. The effect of using a slightly looser definition to remove event overlap for electrons was investigated and found to be negligible.

Line 207: can you give details about what you mean by “less strict”?

The ID is strictly looser to ensure complete overlap removal, but the specifics are not appropriate for the paper.

line 214: "as a correction to simulated events" : unclear what this means. are the event weights modified? if so state so

"Correction to simulated events" is used elsewhere (both in this paper and others). Left as-is

line 216: "the relative isolation variable" is jargon, please rephrase

Done.

line 218: a sentence that starts with a variable is awkward, please rephrase

Done.

line 243: "single lepton trigger without isolation requirement" is jargon, rephrase

This is typically not considered jargon.

line 247: the muon reconstruction coverage goes up to eta = 2.4 so this choice of removing muons in the 2.1 < eta < 2.4 region is bizarre and needs to be justified

This restriction is imposed by the muon trigger, but is not significant enough to warrant a justification in the text.

line 250: "2" -> "two" (spell out the word please)

Done. %ENDCOLOR

Line 250-253: Repetition of “Events are required”

Changed.

line 256: "combination"? do you mean vectorial sum?

Correct. This has been changed.

line 259: "optimized separately" : how were they optimized?

Details of the optimization procedure are too detailed for the paper. The optimzation was performed at multiple masses and done for both discovery significance and exclusion limit.

caption to table 1: 250 and 350 are missing GeV

Fixed.

table 1: please be consistent with the significant digits on the yields in the table. (why are the yields rounded to integers?)

All values are rounded to the nearest integer, so significant figures are consistent. The event counts are large enough that tenths of events are not necessary to report.

table 1: please rename "final selection" (what is final about it?). maybe "full selection"

Changed to full selection.

table 1: would be clearer if instead of "VV" you wrote "WW, WZ, ZZ" (if that's what is intended)

Done.

table 2: the significant digits are inconsistent also in this table

All values are rounded to the nearest integer, so significant figures are consistent.

table 2: The total background uncertainty is very small. How is this justified?

We noticed that there was an inconsistency with the semileptonic table, as there the pure normalization (but not shape) uncertainties were included. The hadronic table has been updated.

Table 2: Please also quote the selection efficiency for different signal scenarios

This information is not appropriate for this table.

table 2: "QCD" is the theory of strong interactions, not the name of a background. rename to multijets, or something like that

Line 273: The working point for b-tagging (apart from the top jet) is nowhere defined

This definition is given in the b-jet identification section.

section 5.1.1: the defined control region is kinematically very far from the signal region: how do you ensure that the results found here also apply to the signal region? is any extrapolation in pt done?

We use two different control regions, with different modifications from the signal selection. We find that the results agree in both control regions, and this gives us confidence that the results can be safely applied to the signal region. Note that no correction is derived in these regions, the effect of the existing top pt reweighting procedure is simply checked.

lines 307 and following: the description of the method is very confusing, please rephrase

A citation for the method is added to help clarify.

line 311: if the results are confirmed by the analytical solution, why not just use the analytical solution instead of the approximate one?

The two yield identical results so there is no difference.

line 313: "we require that the total number of events is unaffected by scale factors": this sentence is obscure. what did you mean? a scale factor in general does modify the event weight and therefore does affect the number of events. what did you mean?

The simultaneous application of the heavy and light flavor scale factors does not affect the number of events (they change number of heavy and light flavor events in a fully correlated way).

lines 314-315: what b-tagging scale factors are you talking about? the ones derived in the above? of the central b-tagging scale factors?

Correct. No other b tagging scale factors have been introduced so it seems unnecessary to clarify this in the text.

line 317 and below: The whole section seems to confuse the usage of “event”, “top candidate”, “b candidate” and “jet”. Events don’t have an n-subjettiness, jets do. Please make sure this is consistent.

line 318: "then": please state what is done in the analysis without stressing the timeline

Removed "then.

line 320: "in order to estimate the average b-tagging rate for multijet events we define alternate t-tagging criteria": sounds confusing (why is the b-tagging affected by t-tagging?) I assume this means that you enlarge the data sample by defining a looser set of cuts. if this is the case please rephrase to state it

The definition serves to define an orthogonal control region. This is clarified in the text.

Line 320 (and below): The section is unclear on how multijet is estimated and from which region the distribution is taken. A figure to illustrate these regions would be highly recommended.

line 320: "an altered" -> "altered"

"An" is necessary here.

line 324: "(less than 1%)": the parentheses are awkward, please rephrase without parentheses

Done.

line 325: "parton flavor distribution" -> "parton flavor composition" (?)

Changed

line 326+1: remove "then"

Done.

line 326+1: "pt": what pt?

b candidate jet. Clarified.

line 326+1: remove the names of the regions (low, transition, high)

Labels are refered to later. Left as-is.

line 326+4: remove "then"

Done.

equation 5: the chosen functional form is bizarre. does this mean that the b-tagging rate is discontinuous at 500 gev? how is this physical?

The b tagging rate is continuous at the boundary (the derivative is not continuous).

equation 5: if x=pt why not write pt?

Changed.

equation 5: add period (.) at the end of the equation

Done.

line 328: do we have enough statistics to appreciate a 50 gev difference from 500 gev to 550 gev in the btagging rate in the forward region? can't we choose the same "a" value for all regions?

Likely the same value would work here, but the shape at high eta has a more pronounced high pt upward trend due to the interplay between b flavor fraction and light mistag fractions in the two regions, so 550 is a slightly better bifurcation point..

line 333: do the alternate functional forms fit the data points? if the points can be fit by a constant, a second degree and third degree polynomials, then we probably don't have the statistical power to distinguish between 500 and 550 gev (see above). if the data points cannot be fit by the alternate functional forms then these tests are not a very significant estimate of the uncertainty

The alternate forms mostly fit the data points. The constant is included as well which contains too few parameters to fully distinguish the functional form, but a constant is a common analysis choice for this type of background estimate (ie ABCD method) therefore it is included here to account for the subjective choice of fit and ensure that the uncertainty band is not optimistic.

line 337: "pass and fail templates" is jargon

Clarified

lines 336-340 are very confusing and it's not clear what is being done

Tried to clarify things. Basically a correlation between b candidate mass and the b discriminator is taken into account when estimating the data driven QCD. The actual effect on the Mtb variable is of course very small but we still like to include it such that we can model all kinematic distributions as well.

line 341: How is this orthogonal? There really needs to be a figure or table with the different event selections

We state that the subjet b tagging selection is inverted, which ensures othogonality

line 345: "small residual discrepancy" please quantify

It is a small shape dependent discrepancy and has a magnitude of about ~5% at points but the total normalization change is negligible

line 346: "final correction" remove final

Done.

line 348: "the following closure test": following what? please rephrase

Done. Ie the correction closes in the control region

line 355: "taken from data": how? if using the above defined method please write that

Done. Right, we just use the same method here.

line 359: consider removing "for both analyses"

Done.

line 363: "uncertainties... luminosity": this sentence reads awkwardly, please rephrase. suggestions: use a different verb instead of "include", "the measured luminosity" -> "the uncertainty on the luminosity measurement"

Done.

line 366: "approx. 5%": please either remove "approx." or replace it with a wiggle

Done.

lines 367-370: this paragraph makes little sense. why are we using different electron IDs in the two channels in the first place? and why does this determine the fact that the hadronic channel has only muon-related uncertainties? doesnt the hadronic channel veto both muons and electrons?

The difference in ID is motivated by technical difference in frameworks (see above). Since the hadronic analysis uses a looser identification for the lepton veto, it is not appropriate to include the same electron ID systematic from the leptonic analysis.

line 379: "+-1sigma" is probably jargon, why not simply say that you vary the scale factors within their uncertainties?

Done.

line 380: "are all included": included where? the chosen verb sounds awkward

Changed,

line 386: "from parton distribution functions" -> "from the choice of the parton distribution functions"

Done.

lines 386-387: "we evaluate the RMS of... as the +-1sigma uncertainties": obscure sentence. what is being done here? please rephrase

Reference added for clarity.

line 390: "is included as a +1sigma variation": awkward way of stating this. do we need to specify the sign of the sigma? can't you just state like you did in line 297 that the difference wrt the uncorrected distribution is taken as an uncertainty?

Done.

line 391: "the uncertainty in" -> "the uncertainty on"

Done.

line 393: "uncertainties on the multijet background are also included" vague sentence. what uncertainties? how large are they? please refrain from using the verb "include" so much.

Clarified.

line 398: "and factorization" -> "and from the choice of the factorization" ?

This phrasing is used only to label the uncertainty, so not necessary. Left as-is.

line 401: "the b tagging scale factor": which factor? is it only one factor?

"scale factor" removed.

line 404: "standard model" either use "SM" or capitalize the S to uniform with the introduction

Changed.

Line 412: uncertainies->uncertainties

Done.

Line 416-417: remove the equal signs

Done.

Line 423: Can you state again the confidence level?

Cumbersome to insert into this statement, and it is given multiple times surrounding it. Left as-is.

table 3: is MC statistics not included as an uncertainty?

Yes, table caption changed to clarify that these are only systematic uncertainties.

table 3: why do all of the uncertainties have +-X% but the muon and electron ID uncertainties don't have the +-?

Fixed.

table 3: the "e" in the single lepton trigger line should not be italic

Fixed.

table 3: some uncertainties are capitalized, some aren’t. please uniform

Done.

table 3: some symbols are never defined (sigma_mb, x)

Corrected.

table 3: missing 1 from +-1sigma in the light quark mistagging line

Fixed.

table 3: are all uncertainties reported in the table described in the text? for example: "b candidate mass". also: some names are rather obscure: what is "average b-tagging rate fit"? how is it different from "b-tagging"? what is "alternate functional forms"? in general i feel that the table is rather confusing and it could be skipped, moving the relevant information in the text.

These are briefly discussed in the systematics section and also in the multijet background estimate section. In our experience (from the last iteration of this result) the full account of the multijet uncertainty is too much detail for a paper.%ENDCOLOR

Table 3: The “1 sigma” mentioned for many uncertainties is useless unless quantified how large a typical/maximal 1 sigma variation is for a given uncertainty.

This information is too detailed for this table and paper. Left as-is.

Table3: The AN also contained a chart of pulls. With less entries than table 3. How does the final pull distribution look?

The chart of pulls is abbreviated to show a subset of pulls which are significant. The pull distributions in the AN are complete.

section 7: it is unclear if the analysis is a shape analysis or a counting experiment. if it's a shape analysis how was the signal shape interpolated between the generated mass points?

The analysis is shape-based. No interpolation is done for signals, samples are generated at 100 GeV intervals. This information is now in the modeling section.

line 408: remove the reference to the theta package (ie end the sentence after the word "prior")

This package needs to be cited.

line 413: "lite" -> "light"?

"Lite" is correct.

line 420: are events with Mtb > 4000 GeV discarded? if so, why?

No, the last bin extends past 4000 GeV, this is clarified.

figure 1: from these plots it is hard to understand if the N(b)=2 category adds any sensitivity at all. does it? if so please provide proof. for example showing a signal that stands above the background uncertainty

The N(b)=2 category actually provides most of the sensitivity.

Figure 1: Remove “The”. Reduce the range of the ratio axes to 1.5. Make the signal histograms more visible

We could not find "The", the axes range we think makes sense here especially if it is to be consistent with Fig 2%ENDCOLOR

The signal histogram lines have been slightly thickened%ENDCOLOR

line 431: "this represents a significant improvement over the results from the individual analyses": is this true? from what stated in line 423 it seems that the leptonic limit is the same as the combined one

The combination improves on the purely leptonic result by about 150 GeV. %ENDCOLOR

line 430: "the observed... is 2.5 (2.4) TeV ": awkward sentence. please rephrase in terms of "masses below X are excluded at 95% confidence level". also: from figure 4 the observed limit and the expected one are spot-on between 2.2 and 2.6 tev: how does this lead to a 100 GeV difference between the expected and observed limits, as stated in the text?

Rephrased. Zooming in on the plot you can see the observed limit dipping below the expected limit, yielding the 100 GeV difference.

figure 2: why does the background prediction suddenly rise around 2200 gev? if it's a statistical fluctuation is this uncertainty correctly propagated to the datacards?

This is a result of the wider bin widths starting here.

figure 2 caption: "of the data set used" sounds awkward, please rephrase

Changed.

Figure 2: Remove “The”

Done.

references 1-10 and 35 and 46 are missing the arxiv

ref 1: why is the title all caps?

Fixed.

refs 1, 6, 7, 9, 14, 28, 35, 46: please remove the number (eg no.1)

Fixed.

Ref 27: Title missing

Fixed.

refs 39 and 40 are incomplete

Fixed.

ref 45 is a website and cannot be cited in a paper

This citation is the correct citation for theta and is used by other CMS papers. Left as-is.

Topic attachments
I Attachment History Action Size Date Who Comment
PNGpng bjetmass.png r1 manage 22.5 K 2017-04-24 - 21:00 DylanSheldonRankin  
PNGpng bpt.png r1 manage 22.9 K 2017-04-24 - 21:00 DylanSheldonRankin  
Edit | Attach | Watch | Print version | History: r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r1 - 2017-04-24 - DylanSheldonRankin
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Sandbox All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright &© 2008-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback