Conclusions

The update results on $F_2$ obtained by this analysis are consistent with the those presented in 1996 even if they are systematically lower due to the overestimation of the photoproduction background. They are in agreement with official ZEUS results, except for few bins in the highest $y$ region, although the discrepancy is within the quoted uncertainty. After all cross checks the only difference between this and the other analysis remains that due to the different unfolding and uncertainty evaluation methods. The agreement can be also seen from the comparison (figure 5) with the NO QCD fit, made using the the published data. In some bins the agreement to the fit is even better than that of the published data. Finally, it is worth noticing that this analysis allows to evaluate the covariance matrix of the data points, which it is not trivial (see table 2) and it can be of crucial importance to extract precise QCD informations from the data.