Hi, I have just figured out that that the intervals indicated by twogtp are not 95% confidence intervals. We used the twogtp confidence intervals in our CG2010 paper, believing they are 95%. But they are in fact standard deviation. They must be multiplied by 1.96 to get 95% confidence. A reviewer of the journal version of our paper correctly pointed out this mistake.
I thought I'd share this with the list. I would not be surprised if we are not the only ones who believed these intervals are 95%. Rémi _______________________________________________ Computer-go mailing list [email protected] http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
