1) NPE v2 in 200, 62, 39 GeV - Daniel
The PAs and GPC are concerned with purity estimation in certian momentum or transverse momentum regions where there could be significant hadron contamination, and therefore propose to not show any of the v2 data points in these regions. Their main concern is that in order to fit the inclusive electron nSigmadEdx distribution, the hadron yields have to be constrained based on interpolating hadron yields from other momentum regions. The correct parameterization to use for such an interpolation is unclear. Moreover, the eID for pT>1 GeV/c uses EMC info and thus reduce the hadron yield. How to include such EMC eID efficiency in interpolating the hadron yields has not been done. 

Suggestion (Zhenyu): plot the hadron yield vs p and show how the interpolation is being done. 
Suggestion (Xin): on the physics plots: compare data with more recent v2 calculations 
Suggestion for future NPE analysis (Daniel): use smaller p bins to allow more reliable interpolation

2) Combinatorial Background Method study - Wangmei
Conclusion: if n_pos=n_neg, arithmetic mean of N++ and N-- is an unbiased estimator for N+-, but geometric mean can underestimate N+- if N+ and N- is small. If n_pos!=n_neg, arithmetic mean could overestimate N+-.
Suggestion (Rosi) there are proposed ways to treat zeros for geometric mean calculation. Check if this help with the underestimation of geometric mean. 
http://www.wwdmag.com/channel/casestudies/handling-zeros-geometric-mean-calculation