I have a question about bias/variance trade-off for different competing models. Say one has estimated model A and model B and calculated their respective train and test error. How does one yield an indication about which of the models has a higher bias? Could one for this just look at the training error (higher training error = higher bias)?
Moreover, for an idea about the variance of the model one could compare the error of training vs. test. What if both models yield a lower test than training error? Does then the model with the even higher difference in error to the training set have a lower variance?
