ME-PS Matching

MATRIX ELEMENT-PARTON SHOWER MATCHING

Updated 09/18/03
Problem with Underlying Event fixed. Events are now hadronized.

Samples of (e+ nu_e)+partons event samples that are matched together using the Pythia parton shower
... at the hadron level (including the underlying event model of Pythia with pi0's decayed, etc.)
... written in a platform-independent, compressed, format using MCFIO with STDHEP
... these are unweighted events, but some of the samples correspond to different luminosities
... the parton level events were generated with MadGraph II with charm as the heaviest flavor

A listing of the directory yields: 72342.W+0run1.1.io 72342.W+0run1.2.io 72342.W+0run1.3.io 72342.W+0run1.4.io 72342.W+0run1.5.io 72342.W+0run1.6.io 72342.W+0run2.1.io 72342.W+0run2.2.io
72342.W+0run2.3.io 72342.W+0run2.4.io 72342.W+0run2.5.io 72342.W+0run2.6.io 72342.W+0run2.7.io 72342.W+0run2.8.io 72342.W+0run3.1.io 72342.W+0run3.2.io
72342.W+0run3.3.io 72342.W+0run3.4.io 72342.W+0run3.5.io 72342.W+0run3.6.io 72342.W+0run3.7.io 72342.W+0run3.8.io 72342.W+0run3.9.io 72342.W+1run1.1.io
72342.W+1run1.2.io 72342.W+1run1.3.io 72342.W+1run4.1.io 72342.W+1run4.2.io 72342.W+1run4.3.io 72342.W+1run4.4.io 72342.W+1run7.1.io 72342.W+1run7.2.io
72342.W+1run7.3.io 72342.W+1run7.4.io 72342.W+2run1.1.io 72342.W+2run1.2.io 72342.W+2run1.3.io 72342.W+2run4.1.io 72342.W+2run4.2.io 72342.W+2run4.3.io
72342.W+2run4.4.io 72342.W+2run7.1.io 72342.W+2run7.2.io 72342.W+2run7.3.io 72342.W+2run7.4.io 72342.W+3run1.1.io 72342.W+3run1.2.io 72342.W+3run4.1.io
72342.W+3run4.2.io 72342.W+3run4.3.io 72342.W+3run7.1.io 72342.W+3run7.2.io 72342.W+3run7.3.io 72342.W+4run1.1.io 72342.W+4run4.1.io 72342.W+4run4.2.io
72342.W+4run7.1.io 72342.W+4run7.2.io wluclusm.tar.gz
These are data files in the MCFIO format and one extra file that gives information about them
The first big number is a batch job number
The "W+n" (where n=0,1,2,3,4) represents the "seed" process whereas the "runX" gives some information about the cuts used for the matching process

Finally, there is an extra integer before ".io", because of some limitation on file sizes to <2 GB.
As an example, the following group of files represents one "set" of events to add together:

72342.W+0run1.1.io, 72342.W+0run1.2.io, 72342.W+0run1.3.io, W+0run1.4.io ... .6.io

It is key that events generated using the same cut are put together at the end, so these events are related to:

The events are at the hadron level, so they are ready to go through a detector simulation
You have to read the files -- an example can be found in the /examples directory of any stdhep version that has been installed using ups/upd
The events are unweighted, but different sets correspond to different luminosities. There is a table below. So, when using the samples W+0run1.x.io through W+4run1.x.io, the total number of events in all the W+0run1.x.io data files represent 733.5 pb, whereas the events in W+4run1.x.io represent only 7.6 pb.
I don't know the exact luminosities without looking through the files (on my to-do list), but I can estimate based on the maximum size of each *.io file. For example, for the cut 10 GeV, the number of events in each data set are roughly: 260k, 100k, 100k, 75k, 50k. Thus, they represent 355, 360, 1234, 3866, 6579 pb^-1 of luminosity respectively (this is just the number of events in each sample divided by the cross section of each sample). The lowest luminosity -- 355 pbi -- sets the limit on how many of the other events to use. Process all of the W+0 events, 355/360=.987 of the W+1 events, etc., all the way down to 355/6579=.054 of the W+4 events. Given my estimate of the total number of W+4 events, this would translate into using about 2700 of them.

The total cross section is different for each cut, and it is smaller than what you expect at NNLO = 1365 pb. The default Pythia cross section is 968 pb. Note, the collider energy is 1960 GeV.