I have seen a lot of research around the Black-Litterman approach and I think theoretically, it is a nice framework. However, it appears that its main strength is from a practitioner's point of view, that it starts with a "reasonable" starting portfolio and "tweaks" it to incorporate investment views. It appears to me that it is most beneficial to portfolio managers with less expertise in the technical and quantitative approaches.
My question is that is this approach used at all by the quantitative managers? Say for example for statistical arbitrage managers who usually have their own sophisticated portfolio construction frameworks?
If so, what is the market equilibrium / starting portfolio. How do you incorporate correlated views, and perhaps most interesting of all, how do you go about estimating the expected returns for the views.
Given that the role of the quantitative hedge funds in the market has been increasing, it's interesting that there is not much research done in this area (please correct me if I'm wrong). Is it because in practice Black-Litterman is not superior to typical approaches for quants?