<HTML>
<HEAD>
<TITLE>Steps to finalizing the pi-p E results</TITLE>
</HEAD>
<BODY>
<FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'>Peng, Dao and Tsuneo,<BR>
<BR>
It would seem we are quite close to finalizing the three analyses of E in the pi-p channel. Let me summarize the proposals from this morning’s g14 meeting:<BR>
(A)<BR>
</SPAN></FONT><UL><LI><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'>For this channel, I propose we plan to report the combined results of periods S1 + S2 + G2a + G2b;<BR>
</SPAN></FONT></UL><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> (I personally think S3 is good as well, but there have been some concerns and it’s too small a set to argue over.)<BR>
</SPAN></FONT><UL><LI><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'>The BDT, KinFit and Empty-subt analyses can ultimately be combined, taking into account the correlations as I showed in the Feb 4 meeting;<BR>
</SPAN></FONT></UL><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> Before combining, we should make a final comparison of the three analyses.<BR>
<BR>
</SPAN></FONT><UL><LI><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'>we will take the limit where the missing spectator momentum => 0; this will be done by multiplying the results by Dao’s correction factor;<BR>
</SPAN></FONT></UL><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> - the BDT and KinFit analyses inherently suppress physics back ground, so that data can be retained out to |p(missing)|< 0.2 GeV/c.<BR>
for this range, the correction factor is 1.08. (Dao will send around the value from his most recent fits.)<BR>
- the Empty-subtraction method is more susceptible to physics background leaking in at higher missing momentum, so that analysis<BR>
will include data only out to |p(missing)|< 0.1 GeV/c. For that case, the correction factor is 1.03. (Dao will send around the most recent value).<BR>
</SPAN></FONT><UL><LI><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'>the next step is to put together comparisons of results from each analysis. (We have looked at partials of this already, but only S1 + G2b and without the P(missing)=>0 correction factor.) I suggest we assemble 2 sets of data, each including the P(missing)=>0 correction factors:<BR>
</SPAN></FONT></UL><FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'> (a) g2a alone, and <BR>
(b) the full S1 + S2 + G2a + G2b.<BR>
<BR>
=> Dao and Tsuneo, please send these to Peng who will assemble the tables.<BR>
<BR>
(B)<BR>
Another comparison that is already underway, and necessary to document in the Analysis Note, is a summary of the cuts used in the three analyses.<BR>
Peng, would you please make a first pass at constructing a table.<BR>
<BR>
Thanks to the three of you for all the hard work,<BR>
Andy</SPAN></FONT>
</BODY>
</HTML>