Compute composite loss value
loss2(y, f, weights, cfun, dfun, s, delta=0.0001)response variable values
linear predictor values of y. If f is predicted response of model, use function loss3 instead
observation weights, same length as y
integer from 1-8, concave function as in ccglm_fit
integer from 1-7, convex function as in ccglm_fit
tuning parameter of cfun. s > 0 and can be equal to 0 for cfun="tcave".
a small positive number provided by user only if cfun="gcave" and 0 < s <1
Weighted loss values
An internal function. For large s values, the loss can be 0 with cfun=2,3,4, or "acave", "bcave", "ccave".
Zhu Wang (2020) Unified Robust Estimation via the COCO, arXiv e-prints, https://arxiv.org/abs/2010.02848