what do we need to know about this? here is my summary, feel free to add to it or correct it ~ similar to a mean variance optimization, except inputs are derived from assuming markets are in equilibrium and using a global market index to calculate expected returns ~ uses the inputs from the index as a starting point and then the analyst can tweak inputs to their own assumptions ~ less bias than regular mean-variance optimization
Overall pretty good summary. Here is mine: 1) it is mean-variance optimization 2) equilibrium weights = market weights -> derive expected returns Starting point is pretty well diversified and stable overtime because market weights are pretty well-balanced and relatively stable. 3) add opinions about expected returns with confidence levels (covar matrrix) -> calculate optimal allocation that incorporates opinions.
I am just going to print out your sumary and keep in my pocket to study the night before the exam 
maratikus Wrote: ------------------------------------------------------- > Overall pretty good summary. Here is mine: > thanks. i remember u from L2 boards. thanks for all your help over the years
mike0021 Wrote: ------------------------------------------------------- > maratikus Wrote: > -------------------------------------------------- > ----- > > Overall pretty good summary. Here is mine: > > > > thanks. i remember u from L2 boards. thanks for > all your help over the years You are welcome. Thanks for your input as well. By the way I realized that it is better to explicitly differentiate B-L approach from Classic mean-variance optimization just as you suggested because of the weaknesses of the classic mean-variance optimization B-L model was developed to overcome.
This is amazing guys, if we can do this kind of thing for all the difficult models/cocepts there in lvl3 , it would be awesome.