Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 32 to 32  
Added:  
> > 
Weights for v15.5.1We use weights from this website: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TopReferences10TeV. Note that the website recommendation is to multiply the crosssection by a factor that includes the kfactor as well as other factors.  
Table of WeightsThe following tables list the properties of the MC samples used in the Single Top analysis. The weights for the single top channels (s, t, and Wt) and ttbar channels were calculated using the NLO cross section. 
Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 33 to 33  
Table of Weights  
Changed:  
< <  The following table lists the properties of the MC samples used in the Single Top analysis. The weights for the single top channels (s, t, and Wt) and ttbar channels were calculated using the NLO cross section. In the 121403 MC version, run 1206, the W+Jets weights include branching ratio, alpgen matching efficiency and filtering efficiency.  
> > 
The following tables list the properties of the MC samples used in the Single Top analysis. The weights for the single top channels (s, t, and Wt) and ttbar channels were calculated using the NLO cross section.
For v13030, the column σ shows σ multiplied by the branching ratio.
Table for v13030
 
Table for Version 121403, weights calculated for 100pb1

Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 19 to 19  
W_{NLO} = W_{} (σ_{NLO} / σ_{LO} ) = σ_{NLO}L/N_{0}  
Added:  
> > 
Branching RatioIf the MC samples were genereated to produce a certain final state, for example ttbar > dileptons, then the branching ratio, BR, for that final state must be included in the weight calculation; W = BRσL/N_{0}  
Sources of informationThe information needed to calculate event weights were collected from the following places:  
Line: 61 to 66  
Table for Version 1213  
Changed:  
< < 
 
> > 
 

Line: 1 to 1  

Event Weights for Monte Carlo SamplesIntroduction to Event Weights  
Changed:  
< <  When a Monte Carlo (MC) sample is generated the size of the sample is determined by requesting the generation of a certain number of events. The cross section of the sample is a fixed quantity dependent on the process generated. Since number of events, sample size, and luminosity are related according to &sigma = L/N_{Events}, the luminosity of a MC sample varies according to the number of events generated and the cross section of the process.  
> >  When a Monte Carlo (MC) sample is generated the size of the sample is determined by requesting the generation of a certain number of events. The cross section of the sample is a fixed quantity dependent on the process generated. Since number of events, sample size, and luminosity are related according to σ = L/N_{Events}, the luminosity of a MC sample varies according to the number of events generated and the cross section of the process.  
The amount of data collected in an experiment is expressed in terms of luminosity since this quantity can be determined from known properties of the colliding beams. In order to perform a simulation of the data at given luminosity, it is necessary to weight the MC sample to correspond to this luminosity. This allows, for example, to predict the number of top events that will be produced in an amount of data corresponding to that luminosity. 
Line: 1 to 1  

Event Weights for Monte Carlo SamplesIntroduction to Event WeightsWhen a Monte Carlo (MC) sample is generated the size of the sample is determined by requesting the generation of a certain number of events. The cross section of the sample is a fixed quantity dependent on the process generated. Since number of events, sample size, and luminosity are related according to &sigma=L/N_{Events}, the luminosity of a MC sample varies according to the number of events generated and the cross section of the process.  
Changed:  
< <  The amount of data collected in an experiemnt is expressed in terms of luminosity since this quantity can be determined from known properties of the colliding beams. In order to perform a simulation of the data at given luminosity, it is necessary to weight the MC sample to correspond to this luminosity. This allows, for example, to predict the number of top events that will be produced in an amount of data corresponding to that luminosity.  
> >  The amount of data collected in an experiment is expressed in terms of luminosity since this quantity can be determined from known properties of the colliding beams. In order to perform a simulation of the data at given luminosity, it is necessary to weight the MC sample to correspond to this luminosity. This allows, for example, to predict the number of top events that will be produced in an amount of data corresponding to that luminosity.  
The weighting of the MC can be expressed as N = WN_{0}, where N is the number of weighted events, N_{0} is the number of events in the original sample, and W is the weight. The weight must be a ratio of the desired luminsoity to the original luminosity of the sample, W = L/L_{0}. Expressing the original luminosity in terms of cross section and number of events, 
Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 28 to 28  
Table of Weights  
Changed:  
< <  The following table lists the properties of the MC samples used in the Single Top analysis. The weights for the single top channels (s, t, and Wt) and ttbar channels were calculated using the NLO cross section. In the 121403 MC version, the W+Jets weights include branching ratio, alpgen matching efficiency and filtering efficiency.  
> >  The following table lists the properties of the MC samples used in the Single Top analysis. The weights for the single top channels (s, t, and Wt) and ttbar channels were calculated using the NLO cross section. In the 121403 MC version, run 1206, the W+Jets weights include branching ratio, alpgen matching efficiency and filtering efficiency.  
Table for Version 121403, weights calculated for 100pb1

Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 30 to 30  
Table of WeightsThe following table lists the properties of the MC samples used in the Single Top analysis. The weights for the single top channels (s, t, and Wt) and ttbar channels were calculated using the NLO cross section. In the 121403 MC version, the W+Jets weights include branching ratio, alpgen matching efficiency and filtering efficiency.Table for Version 121403, weights calculated for 100pb1  
Changed:  
< < 
 
> > 
 
 
Line: 72 to 71  
Deleted:  
< <   PatRyan  19 Feb 2007,  JennyHolzbauer  01 Aug 2007  
\ No newline at end of file  
Added:  
> >   PatRyan  19 Feb 2007,  JennyHolzbauer  01 Aug 2007 JennyHolzbauer  30 Jan 2008  
\ No newline at end of file 
Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 29 to 29  
Table of WeightsThe following table lists the properties of the MC samples used in the Single Top analysis. The weights for the single top channels (s, t, and Wt) and ttbar channels were calculated using the NLO cross section. In the 121403 MC version, the W+Jets weights include branching ratio, alpgen matching efficiency and filtering efficiency.  
Changed:  
< < 
Table for Version 121403
 
> > 
Table for Version 121403, weights calculated for 100pb1
 

Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 28 to 28  
Table of Weights  
Changed:  
< <  The following table lists the properties of the MC samples used in the Single Top analysis. The weights for the single top channels (s, t, and Wt) and ttbar channels were calculated using the NLO cross section.  
> > 
The following table lists the properties of the MC samples used in the Single Top analysis. The weights for the single top channels (s, t, and Wt) and ttbar channels were calculated using the NLO cross section. In the 121403 MC version, the W+Jets weights include branching ratio, alpgen matching efficiency and filtering efficiency.
Table for Version 121403
 
Added:  
> > 
Table for Version 1214  

Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 48 to 48  
Table for Version 1213  
Changed:  
< < 
 
> > 
 
Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 51 to 51  
 
Changed:  
< < 
 
> > 
 

Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 46 to 46  
 
Added:  
> >  
Table for Version 1213
 
Changed:  
< < 
 
> > 
 

Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 47 to 47  
Table for Version 1213  
Changed:  
< < 
 
> > 
 
 
Deleted:  
< <   PatRyan  19 Feb 2007  JennyHolzbauer  01 Aug 2007  
\ No newline at end of file  
Added:  
> >   PatRyan  19 Feb 2007,  JennyHolzbauer  01 Aug 2007  
\ No newline at end of file 
Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 46 to 46  
 
Added:  
> > 
Table for Version 1213
 
 PatRyan  19 Feb 2007 \ No newline at end of file  
Added:  
> >   JennyHolzbauer  01 Aug 2007  
\ No newline at end of file 
Line: 1 to 1  

Event Weights for Monte Carlo SamplesIntroduction to Event Weights  
Changed:  
< <  The general formula for event weights is W = σL/N_{Events}, where σ is the cross section of the generated process, L is the generated luminosity of the MC sample, and N_{Events} is the number of generated events.  
> >  When a Monte Carlo (MC) sample is generated the size of the sample is determined by requesting the generation of a certain number of events. The cross section of the sample is a fixed quantity dependent on the process generated. Since number of events, sample size, and luminosity are related according to &sigma=L/N_{Events}, the luminosity of a MC sample varies according to the number of events generated and the cross section of the process.  
Changed:  
< <  The MC samples we are using were generated with the correct weights and therefore in the simplest case we would not have to weight our events. However, the weighting of events is desired in the following cases:  
> >  The amount of data collected in an experiemnt is expressed in terms of luminosity since this quantity can be determined from known properties of the colliding beams. In order to perform a simulation of the data at given luminosity, it is necessary to weight the MC sample to correspond to this luminosity. This allows, for example, to predict the number of top events that will be produced in an amount of data corresponding to that luminosity.  
Changed:  
< < 
NLO vs. LOAll the MC samples were genereated using a LO cross section prediction. For the single top processes (s, t, and Wt channels), the cross sections are known to NLO and the weights are scaled by a ratio of NLO to LO cross sections to reflect this.  
> >  The weighting of the MC can be expressed as N = WN_{0}, where N is the number of weighted events, N_{0} is the number of events in the original sample, and W is the weight. The weight must be a ratio of the desired luminsoity to the original luminosity of the sample, W = L/L_{0}. Expressing the original luminosity in terms of cross section and number of events, W = σL/N_{0}.  
Changed:  
< <  W_{new} = (σ_{NLO}/σ_{LO})W.  
> >  All histograms are weighted by multiplying the quantity used to fill the histogram (N, pT, eta, etc) by the event weight. In the MSU analysis code, the actual weighting takes place in the method myTH1F::Fill().  
Changed:  
< < 
Predictions for a certain amount of DataWe would like to determine how many single top events we will see in 1fb^{1} of data. This is achieved by scaling the weight by the ratio of 1fb^{1} to the generated luminosity, L (1fb^{1} = 1000pb^{1}).  
> > 
NLO vs. LOAll the MC samples were genereated using a LO cross section prediction. For the single top (s, t, and Wt channels) ttbar processes, the cross sections are known to NLO. In this case the weight is scaled by the ratio of NLO to LO cross sections  
Changed:  
< <  W_{new} = (1000 pb^{1}/L)W.  
> >  W_{NLO} = W_{} (σ_{NLO} / σ_{LO} ) = σ_{NLO}L/N_{0}  
Sources of information  
Changed:  
< <  The information (cross sections, luminosities, etc) needed to calculate event weights were collected from the following places:  
> >  The information needed to calculate event weights were collected from the following places:  
Line: 27 to 28  
Table of Weights  
Changed:  
< <  The following table lists the properties of the MC samples used in the Single Top analysis. The weight was calculated by the formula W = (1000pb^{1}). In addition, for the single top channels (s, t, and Wt), the weight was scaled by the ratio (σ_{NLO} / σ_{LO}).  
> >  The following table lists the properties of the MC samples used in the Single Top analysis. The weights for the single top channels (s, t, and Wt) and ttbar channels were calculated using the NLO cross section.  
Changed:  
< < 
 
> > 
 

Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 26 to 26  
Added:  
> > 
Table of WeightsThe following table lists the properties of the MC samples used in the Single Top analysis. The weight was calculated by the formula W = (1000pb^{1}). In addition, for the single top channels (s, t, and Wt), the weight was scaled by the ratio (σ_{NLO} / σ_{LO}).
 
 PatRyan  19 Feb 2007 \ No newline at end of file 
Line: 1 to 1  

Event Weights for Monte Carlo Samples  
Line: 18 to 18  
W_{new} = (1000 pb^{1}/L)W.  
Added:  
> > 
Sources of informationThe information (cross sections, luminosities, etc) needed to calculate event weights were collected from the following places:  
 PatRyan  19 Feb 2007 \ No newline at end of file 
Line: 1 to 1  

 
Changed:  
< <  Event Weights info here  
> > 
Event Weights for Monte Carlo SamplesIntroduction to Event WeightsThe general formula for event weights is W = σL/N_{Events}, where σ is the cross section of the generated process, L is the generated luminosity of the MC sample, and N_{Events} is the number of generated events. The MC samples we are using were generated with the correct weights and therefore in the simplest case we would not have to weight our events. However, the weighting of events is desired in the following cases:NLO vs. LOAll the MC samples were genereated using a LO cross section prediction. For the single top processes (s, t, and Wt channels), the cross sections are known to NLO and the weights are scaled by a ratio of NLO to LO cross sections to reflect this. W_{new} = (σ_{NLO}/σ_{LO})W.Predictions for a certain amount of DataWe would like to determine how many single top events we will see in 1fb^{1} of data. This is achieved by scaling the weight by the ratio of 1fb^{1} to the generated luminosity, L (1fb^{1} = 1000pb^{1}). W_{new} = (1000 pb^{1}/L)W.  