search for: ngoodobs

Displaying 1 result from an estimated 1 matches for "ngoodobs".

2002 Feb 27
1
Bug in glm.fit? (PR#1331)
...(paste("No observations informative at iteration", 74 iter)) 75 break 76 } 77 z <- (eta - offset)[good] + (y - mu)[good]/mu.eta.val[good] 78 w <- sqrt((weights[good] * mu.eta.val[good]^2)/variance(mu)[good]) 79 ngoodobs <- as.integer(nobs - sum(!good)) 80 ncols <- as.integer(1) 81 fit <- .Fortran("dqrls", qr = x[good, ] * w, n = as.integer(ngoodobs), 82 p = nvars, y = w * z, ny = ncols, tol = min(1e-07, 83 control$epsilon/1000), coefficients...